Measured performance
Provider & model performance
Measured time-to-first-token, time-to-first-byte, effective throughput, and uptime for every LLM provider and model TrustedRouter routes to — continuously sampled, not vendor-claimed.
Last updated
2026-08-01T20:39:58Z
Continuously sampled from TrustedRouter's monitor regions using a
rolling benchmark set of up to 10,000 samples. Time-to-first-token (TTFT),
time-to-first-byte (TTFB), effective throughput, and success rate come from real
streaming requests, not vendor claims. Completion rate shows the end-user result.
Provider availability excludes failures owned by TrustedRouter, customers, or route
configuration, while still reporting those failures under their actual owner.
Capacity acceptance measures requests not rejected for provider overload. Effective throughput is provider-reported
output tokens divided by complete request time, so buffered delivery cannot inflate
the result. Unsupported route and probe-configuration rows are reported separately
and do not count as provider downtime. No prompt or output content is ever stored.
Providers
Ranked by provider-attributed availability, then p50 time-to-first-token across all of a provider's models (40 providers · 2344 availability samples · 233 sustained throughput samples). Providers below the evidence floor remain visible as warming, but do not receive a rank.
| # | Provider | Models | Completion | Provider availability | Within deadline | Capacity accepted | p50 TTFT | p95 TTFT | Effective throughput | Failure owner | Config excluded | Samples |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | cerebras | 6 | 100.00% | 100.00% | 100.00% | 100.00% | 1143 ms | 4740 ms | 126 tok/s | — | — | 68 |
| 2 | mistral | 8 | 100.00% | 100.00% | 100.00% | 100.00% | 1427 ms | 3089 ms | — | — | — | 66 |
| 3 | openai | 19 | 100.00% | 100.00% | 100.00% | 100.00% | 1443 ms | 4125 ms | 61 tok/s | — | — | 67 |
| 4 | fireworks | 11 | 100.00% | 100.00% | 100.00% | 100.00% | 1464 ms | 4736 ms | 59 tok/s | — | — | 64 |
| 5 | neurometric | 5 | 100.00% | 100.00% | 100.00% | 100.00% | 1510 ms | 3347 ms | — | — | — | 67 |
| 6 | makora | 4 | 100.00% | 100.00% | 100.00% | 100.00% | 1551 ms | 3743 ms | 131 tok/s | — | — | 41 |
| 7 | grok | 4 | 100.00% | 100.00% | 100.00% | 100.00% | 1692 ms | 3662 ms | 44 tok/s | — | — | 61 |
| 8 | inceptron | 3 | 100.00% | 100.00% | 100.00% | 100.00% | 1801 ms | 8693 ms | 157 tok/s | — | 1 probe_config_error |
50 |
| 9 | together | 14 | 100.00% | 100.00% | 100.00% | 100.00% | 1854 ms | 9257 ms | 66 tok/s | — | 14 probe_config_error |
34 |
| 10 | kimi | 11 | 100.00% | 100.00% | 100.00% | 100.00% | 1945 ms | 5564 ms | 26 tok/s | — | — | 58 |
| 11 | deepseek | 3 | 100.00% | 100.00% | 100.00% | 100.00% | 1991 ms | 3650 ms | 57 tok/s | — | — | 64 |
| 12 | minimax | 6 | 100.00% | 100.00% | 100.00% | 100.00% | 2012 ms | 4164 ms | 69 tok/s | — | — | 56 |
| 13 | anthropic | 10 | 100.00% | 100.00% | 100.00% | 100.00% | 2066 ms | 7409 ms | 66 tok/s | — | — | 52 |
| 14 | wafer | 5 | 100.00% | 100.00% | 100.00% | 100.00% | 2081 ms | 4344 ms | 143 tok/s | — | — | 63 |
| 15 | streamlake | 3 | 100.00% | 100.00% | 100.00% | 100.00% | 2090 ms | 4251 ms | 93 tok/s | — | — | 57 |
| 16 | google-vertex | 9 | 100.00% | 100.00% | 100.00% | 100.00% | 2163 ms | 5354 ms | 56 tok/s | — | — | 51 |
| 17 | telnyx | 10 | 100.00% | 100.00% | 100.00% | 100.00% | 2335 ms | 3844 ms | 92 tok/s | — | 2 probe_config_error |
51 |
| 18 | venice | 12 | 100.00% | 100.00% | 100.00% | 100.00% | 2481 ms | 17844 ms | 33 tok/s | — | — | 59 |
| 19 | lightning | 12 | 100.00% | 100.00% | 100.00% | 100.00% | 2530 ms | 4284 ms | 71 tok/s | — | — | 50 |
| 20 | google-ai-studio | 8 | 100.00% | 100.00% | 100.00% | 100.00% | 2748 ms | 4624 ms | 79 tok/s | — | — | 63 |
| 21 | gmi | 5 | 100.00% | 100.00% | 100.00% | 100.00% | 3531 ms | 7035 ms | 39 tok/s | — | — | 67 |
| 22 | thinkingmachines | 2 | 98.53% | 98.53% | 98.53% | 100.00% | 1994 ms | 13097 ms | 38 tok/s | provider 1 |
— | 68 |
| 23 | morph | 8 | 98.46% | 98.46% | 98.46% | 100.00% | 2001 ms | 4740 ms | 42 tok/s | provider 1 |
— | 65 |
| 24 | digitalocean | 21 | 98.44% | 98.44% | 98.44% | 98.44% | 1586 ms | 5351 ms | 28 tok/s | provider 1 |
— | 64 |
| 25 | xiaomi | 3 | 98.41% | 98.41% | 98.41% | 100.00% | 2830 ms | 5979 ms | 35 tok/s | provider 1 |
— | 63 |
| 26 | nebius | 24 | 98.33% | 98.33% | 98.33% | 100.00% | 2047 ms | 9987 ms | 85 tok/s | provider 1 |
— | 60 |
| 27 | siliconflow | 26 | 98.33% | 98.33% | 96.55% | 98.33% | 2517 ms | 5304 ms | 51 tok/s | provider 1 |
4 unsupported_route |
60 |
| 28 | friendli | 7 | 98.28% | 98.28% | 98.28% | 100.00% | 1368 ms | 4244 ms | 94 tok/s | provider 1 |
— | 58 |
| 29 | cloudflare-workers-ai | 18 | 98.18% | 98.18% | 98.18% | 100.00% | 1459 ms | 5087 ms | 37 tok/s | provider 1 |
— | 55 |
| 30 | phala | 2 | 98.11% | 98.11% | 98.11% | 100.00% | 2361 ms | 9099 ms | 49 tok/s | provider 1 |
— | 53 |
| 31 | atlas-cloud | 50 | 96.92% | 96.92% | 96.92% | 100.00% | 1928 ms | 3980 ms | 65 tok/s | provider 2 |
— | 65 |
| 32 | zai | 11 | 96.72% | 96.72% | 95.08% | 100.00% | 2035 ms | 4939 ms | 67 tok/s | provider 2 |
— | 61 |
| 33 | deepinfra | 39 | 96.49% | 96.49% | 96.49% | 96.49% | 1874 ms | 7660 ms | 49 tok/s | provider 2 |
— | 57 |
| 34 | crusoe | 14 | 94.83% | 94.83% | 94.83% | 94.83% | 2009 ms | 4984 ms | 63 tok/s | provider 3 |
— | 58 |
| 35 | alibaba | 39 | 94.12% | 94.12% | 94.12% | 100.00% | 2253 ms | 4123 ms | 40 tok/s | provider 3 |
— | 51 |
| 36 | chutes | 11 | 94.12% | 94.12% | 90.20% | 98.04% | 2460 ms | 15854 ms | 31 tok/s | provider 3 |
1 unsupported_route |
51 |
| 37 | baseten | 11 | 94.03% | 94.03% | 92.54% | 100.00% | 2117 ms | 3543 ms | 72 tok/s | provider 4 |
— | 67 |
| 38 | novita | 54 | 92.19% | 92.19% | 92.19% | 96.88% | 2206 ms | 7456 ms | 37 tok/s | provider 5 |
— | 64 |
| 39 | tinfoil | 5 | 90.16% | 90.16% | 88.52% | 100.00% | 1705 ms | 5241 ms | 84 tok/s | provider 6 |
— | 61 |
| 40 | parasail | 20 | 87.04% | 87.04% | 83.33% | 87.04% | 2730 ms | 5333 ms | 56 tok/s | provider 7 |
1 probe_config_error |
54 |
Models
Qualified models are ranked by provider-attributed availability, then p50 TTFT. Thin availability and throughput measurements stay visible below the ranked set as warming.
| # | Model | Provider | Completion | Provider availability | Within deadline | Capacity accepted | p50 TTFT | p95 TTFT | p50 TTFB | p95 TTFB | Effective throughput | Config excluded | Samples |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | cerebras/gpt-oss-120b | cerebras | 100.00% | 100.00% | 100.00% 15s | 100.00% | 980 ms | 3344 ms | 752 ms | 3344 ms | 377 tok/s warming | — | 23 |
| 2 | openai/gpt-4.1-mini | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1075 ms | 3222 ms | 1075 ms | 3222 ms | 61 tok/s warming | — | 12 |
| 3 | z-ai/glm-4.7 | cerebras | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1143 ms | 4088 ms | 1142 ms | 4088 ms | — | — | 11 |
| 4 | deepseek/deepseek-v4-pro | makora | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1263 ms | 4660 ms | 1262 ms | 4660 ms | 127 tok/s warming | — | 10 |
| 5 | google/gemini-3-flash-preview | google-ai-studio | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1264 ms | 3807 ms | 1264 ms | 3807 ms | — | — | 12 |
| 6 | z-ai/glm-5.2 | friendli | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1268 ms | 3348 ms | 1267 ms | 3348 ms | 94 tok/s warming | — | 14 |
| 7 | mistralai/mistral-medium-3-5 | mistral | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1304 ms | 3089 ms | 1304 ms | 3089 ms | — | — | 11 |
| 8 | google/gemma-4-31b-it | tinfoil | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1360 ms | 3386 ms | 1360 ms | 3386 ms | 205 tok/s warming | — | 18 |
| 9 | kwaipilot/kat-coder-pro-v2 | streamlake | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1406 ms | 4203 ms | 1406 ms | 4203 ms | — | — | 22 |
| 10 | qwen/qwen3-vl-8b-instruct | neurometric | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1412 ms | 3629 ms | 1412 ms | 3629 ms | — | — | 14 |
| 11 | morph/morph-v3-large | morph | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1436 ms | 2902 ms | 1436 ms | 2902 ms | — | — | 11 |
| 12 | deepseek/deepseek-v4-flash | fireworks | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1464 ms | 3670 ms | 1464 ms | 3670 ms | 153 tok/s warming | — | 10 |
| 13 | z-ai/glm-5.1 | wafer | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1470 ms | 3355 ms | 1470 ms | 3355 ms | — | — | 13 |
| 14 | qwen/qwen3-vl-8b-thinking | neurometric | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1510 ms | 3327 ms | 1510 ms | 3327 ms | — | — | 16 |
| 15 | ibm-granite/granite-4.1-8b | neurometric | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1528 ms | 3411 ms | 1528 ms | 3411 ms | — | — | 21 |
| 16 | x-ai/grok-build-0.1 | grok | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1549 ms | 3223 ms | 1549 ms | 3223 ms | — | — | 16 |
| 17 | deepseek/deepseek-v4-flash | makora | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1551 ms | 3278 ms | 1551 ms | 3278 ms | 135 tok/s warming | — | 15 |
| 18 | mistralai/mistral-small-2603 | mistral | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1569 ms | 4039 ms | 1569 ms | 4038 ms | — | — | 10 |
| 19 | x-ai/grok-4.20 | grok | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1600 ms | 3136 ms | 1599 ms | 3136 ms | — | — | 12 |
| 20 | x-ai/grok-4.3 | grok | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1692 ms | 12316 ms | 1692 ms | 12316 ms | — | — | 18 |
| 21 | moonshotai/kimi-k2.7-code | inceptron | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1748 ms | 3471 ms | 1748 ms | 3471 ms | 157 tok/s warming | — | 21 |
| 22 | x-ai/grok-4.5 | grok | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1780 ms | 3752 ms | 1780 ms | 3752 ms | 44 tok/s warming | — | 15 |
| 23 | moonshotai/kimi-k2.6 | inceptron | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1801 ms | 3143 ms | 1801 ms | 3143 ms | — | 1 probe_config_error |
14 |
| 24 | deepseek/deepseek-v4-flash-0731 | deepseek | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1807 ms | 3650 ms | 1806 ms | 3649 ms | 120 tok/s warming | — | 15 |
| 25 | minimax/minimax-m2.1-highspeed | minimax | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1835 ms | 3489 ms | 1835 ms | 3489 ms | — | — | 12 |
| 26 | deepseek/deepseek-v4-flash | deepseek | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1991 ms | 3810 ms | 1991 ms | 3810 ms | 59 tok/s warming | — | 25 |
| 27 | thinkingmachines/inkling-small | thinkingmachines | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1994 ms | 5060 ms | 1994 ms | 5060 ms | 93 tok/s warming | — | 36 |
| 28 | minimax/minimax-m2 | minimax | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2012 ms | 3790 ms | 2012 ms | 3790 ms | — | — | 11 |
| 29 | minimax/minimax-m3 | wafer | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2081 ms | 3743 ms | 2081 ms | 3743 ms | 97 tok/s warming | — | 21 |
| 30 | kwaipilot/kat-coder-air-v2.5 | streamlake | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2090 ms | 4180 ms | 2090 ms | 4180 ms | 93 tok/s warming | — | 19 |
| 31 | moonshotai/kimi-k3 | baseten | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2117 ms | 3800 ms | 2117 ms | 3799 ms | 47 tok/s warming | — | 10 |
| 32 | deepseek/deepseek-v4-pro | deepseek | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2146 ms | 3525 ms | 2146 ms | 3524 ms | 55 tok/s warming | — | 24 |
| 33 | minimax/minimax-m2.7-highspeed | minimax | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2172 ms | 4164 ms | 2172 ms | 4164 ms | — | — | 11 |
| 34 | google/gemini-3.5-flash | google-ai-studio | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2258 ms | 3998 ms | 2258 ms | 3998 ms | — | — | 10 |
| 35 | z-ai/glm-5.2 | phala | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2361 ms | 5124 ms | 2361 ms | 5124 ms | 49 tok/s warming | — | 27 |
| 36 | z-ai/glm-5.2 | wafer | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2426 ms | 5346 ms | 2425 ms | 5346 ms | 190 tok/s warming | — | 14 |
| 37 | xiaomi/mimo-v2.5-pro-ultraspeed | xiaomi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2510 ms | 4485 ms | 2510 ms | 4484 ms | 183 tok/s warming | — | 21 |
| 38 | minimax/minimax-m3 | fireworks | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2555 ms | 4736 ms | 2554 ms | 4736 ms | 73 tok/s warming | — | 10 |
| 39 | kwaipilot/kat-coder-pro-v2.5 | streamlake | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2628 ms | 4528 ms | 2628 ms | 4528 ms | — | — | 16 |
| 40 | z-ai/glm-5.2 | parasail | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2730 ms | 12604 ms | 984 ms | 1880 ms | 62 tok/s warming | — | 13 |
| 41 | z-ai/glm-5.1 | gmi | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2788 ms | 5205 ms | 2787 ms | 5204 ms | — | — | 10 |
| 42 | z-ai/glm-5.2-fast | wafer | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2788 ms | 4812 ms | 2788 ms | 4811 ms | 219 tok/s warming | — | 14 |
| 43 | xiaomi/mimo-v2.5-pro | xiaomi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2830 ms | 7007 ms | 2830 ms | 7007 ms | 35 tok/s warming | — | 27 |
| 44 | google/gemini-3.1-flash-lite | google-ai-studio | 100.00% | 100.00% | 100.00% 15s | 100.00% | 2946 ms | 4139 ms | 2946 ms | 4139 ms | — | — | 12 |
| 45 | google/gemini-2.5-flash | lightning | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3119 ms | 3501 ms | 3118 ms | 3501 ms | — | — | 10 |
| 46 | google/gemma-4-31b-it | cerebras | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3246 ms | 8703 ms | 3246 ms | 8703 ms | 126 tok/s warming | — | 10 |
| 47 | z-ai/glm-5 | gmi | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3313 ms | 24717 ms | 3312 ms | 24717 ms | — | — | 14 |
| 48 | deepseek/deepseek-v4-pro | gmi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3531 ms | 13242 ms | 3531 ms | 13241 ms | 29 tok/s warming | — | 17 |
| 49 | z-ai/glm-5.2 | gmi | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3941 ms | 5872 ms | 3941 ms | 5872 ms | 50 tok/s warming | — | 18 |
| 50 | z-ai/glm-5.2 | inceptron | 100.00% | 100.00% | 100.00% 45s | 100.00% | 6474 ms | 29961 ms | 6474 ms | 29960 ms | — | — | 15 |
| 51 | thinkingmachines/inkling | thinkingmachines | 96.88% | 96.88% | 96.88% 45s | 100.00% | 4105 ms | 26943 ms | 4105 ms | 26943 ms | 38 tok/s warming | — | 32 |
| 52 | moonshotai/kimi-k3 | phala | 96.15% | 96.15% | 96.15% 20s | 100.00% | 2962 ms | 17153 ms | 2961 ms | 17152 ms | — | — | 26 |
| 53 | xiaomi/mimo-v2.5 | xiaomi | 93.33% | 93.33% | 93.33% 20s | 100.00% | 3317 ms | 15555 ms | 3317 ms | 15554 ms | — | — | 15 |
| 54 | qwen/qwen3-235b-a22b-2507 | friendli | 92.31% | 92.31% | 92.31% 20s | 100.00% | 1368 ms | 3697 ms | 1368 ms | 3697 ms | — | — | 13 |
| 55 | moonshotai/kimi-k2.6 | baseten | 90.91% | 90.91% | 90.91% 20s | 100.00% | 2220 ms | 3969 ms | 2220 ms | 3968 ms | 97 tok/s warming | — | 11 |
| 56 | qwen/qwen3.5-397b-a17b | morph | 90.91% | 90.91% | 90.91% 20s | 100.00% | 2289 ms | 4010 ms | 2289 ms | 4010 ms | — | — | 11 |
| 57 | z-ai/glm-4.5 | zai | 90.00% | 90.00% | 90.00% 20s | 100.00% | 2756 ms | 4630 ms | 2756 ms | 4630 ms | — | — | 10 |
| 58 | z-ai/glm-5.2 | tinfoil | 84.62% | 84.62% | 76.92% 45s | 100.00% | 3541 ms | 46057 ms | 3541 ms | 46057 ms | — | — | 13 |
| 59 | meta-llama/llama-3.3-70b-instruct | tinfoil | 83.33% | 83.33% | 83.33% 20s | 100.00% | 1705 ms | 4547 ms | 1704 ms | 4547 ms | 43 tok/s warming | — | 12 |
| — | qwen/qwen3.5-27b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 369 ms | 369 ms | 368 ms | 368 ms | — | — | 1 |
| — | qwen/qwen3.5-9b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 476 ms | 544 ms | 476 ms | 544 ms | — | — | 2 |
| — | deepseek/deepseek-r1-0528 warming | deepinfra | 100.00% | 100.00% | 100.00% 45s | 100.00% | 517 ms | 1838 ms | 516 ms | 1838 ms | — | — | 2 |
| — | qwen/qwen3-30b-a3b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 548 ms | 548 ms | 548 ms | 548 ms | — | — | 1 |
| — | qwen/qwen3-coder-next warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 633 ms | 1358 ms | 632 ms | 1357 ms | — | — | 2 |
| — | minimax/minimax-m2.5 warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 651 ms | 651 ms | 651 ms | 651 ms | — | — | 1 |
| — | deepseek/deepseek-v4-pro warming | fireworks | 100.00% | 100.00% | 100.00% 20s | 100.00% | 667 ms | 1367 ms | 667 ms | 1366 ms | 59 tok/s warming | — | 4 |
| — | qwen/qwen3-next-80b-a3b-instruct warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 692 ms | 4639 ms | 692 ms | 4639 ms | — | — | 2 |
| — | openai/gpt-oss-120b warming | together | 100.00% | 100.00% | 100.00% 20s | 100.00% | 697 ms | 3670 ms | 697 ms | 3670 ms | 129 tok/s warming | — | 2 |
| — | qwen/qwen3.5-35b-a3b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 698 ms | 777 ms | 698 ms | 776 ms | — | — | 2 |
| — | moonshotai/kimi-k2.7-code warming | fireworks | 100.00% | 100.00% | 100.00% 20s | 100.00% | 699 ms | 1727 ms | 698 ms | 1727 ms | 59 tok/s warming | — | 5 |
| — | deepseek/deepseek-v4-flash warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 705 ms | 1017 ms | 705 ms | 1017 ms | 49 tok/s warming | — | 2 |
| — | x-ai/grok-build-0.1 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 722 ms | 1212 ms | 722 ms | 1212 ms | — | — | 2 |
| — | nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 726 ms | 1049 ms | 725 ms | 1049 ms | — | — | 2 |
| — | z-ai/glm-5.2-fast warming | fireworks | 100.00% | 100.00% | 100.00% 45s | 100.00% | 730 ms | 3060 ms | 729 ms | 3060 ms | 55 tok/s warming | — | 4 |
| — | openai/gpt-5.1 warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 735 ms | 3392 ms | 735 ms | 3391 ms | — | — | 2 |
| — | openai/o4-mini warming | openai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 764 ms | 2010 ms | 764 ms | 2009 ms | — | — | 2 |
| — | qwen/qwen3-235b-a22b-thinking-2507 warming | deepinfra | 100.00% | 100.00% | 100.00% 45s | 100.00% | 764 ms | 764 ms | 764 ms | 764 ms | — | — | 1 |
| — | openai/gpt-4o-mini warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 791 ms | 2998 ms | 791 ms | 2998 ms | — | — | 4 |
| — | meta-llama/llama-3.1-8b-instruct warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 807 ms | 807 ms | 806 ms | 806 ms | — | — | 1 |
| — | meta-llama/llama-3.3-70b-instruct-fp8-fast warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 15s | 100.00% | 816 ms | 2624 ms | 816 ms | 2624 ms | — | — | 3 |
| — | morph/morph-v3-fast warming | morph | 100.00% | 100.00% | 100.00% 15s | 100.00% | 817 ms | 3035 ms | 817 ms | 3034 ms | — | — | 9 |
| — | qwen/qwen-mt-plus warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 820 ms | 1766 ms | 820 ms | 1766 ms | — | — | 2 |
| — | openai/gpt-oss-20b warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 830 ms | 3016 ms | 829 ms | 3015 ms | — | — | 4 |
| — | google/gemma-2b-it warming | telnyx | 100.00% | 100.00% | 100.00% 20s | 100.00% | 845 ms | 1636 ms | 845 ms | 1635 ms | — | — | 5 |
| — | deepseek/deepseek-v4-flash warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 882 ms | 882 ms | 882 ms | 882 ms | 37 tok/s warming | — | 1 |
| — | qwen/qwen3-vl-30b-a3b-instruct warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 892 ms | 3406 ms | 892 ms | 3406 ms | — | — | 2 |
| — | google/gemma-4-12b-it warming | neurometric | 100.00% | 100.00% | 100.00% 20s | 100.00% | 893 ms | 3218 ms | 893 ms | 3218 ms | — | — | 8 |
| — | minimax/minimax-m2.5 warming | friendli | 100.00% | 100.00% | 100.00% 20s | 100.00% | 901 ms | 4024 ms | 901 ms | 4024 ms | — | — | 6 |
| — | qwen/qwen3-coder-30b-a3b-instruct warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 932 ms | 1733 ms | 932 ms | 1733 ms | — | — | 2 |
| — | openai/gpt-4.1 warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 942 ms | 1382 ms | 942 ms | 1382 ms | — | — | 3 |
| — | nvidia/nemotron-3-nano-omni warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 962 ms | 3394 ms | 961 ms | 3394 ms | — | — | 4 |
| — | qwen/qwen3-30b-a3b-fp8 warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 963 ms | 2926 ms | 963 ms | 2926 ms | — | — | 4 |
| — | qwen/qwen3.5-397b-a17b warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 965 ms | 1374 ms | 965 ms | 1374 ms | — | — | 2 |
| — | openai/gpt-4o warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 967 ms | 1877 ms | 967 ms | 1877 ms | — | — | 4 |
| — | mistralai/ministral-14b-2512 warming | mistral | 100.00% | 100.00% | 100.00% 20s | 100.00% | 977 ms | 3025 ms | 977 ms | 3025 ms | — | — | 7 |
| — | moonshotai/kimi-k2.6 warming | together | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1000 ms | 1158 ms | 1000 ms | 1158 ms | — | — | 3 |
| — | z-ai/glm-4.7 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1003 ms | 1066 ms | 1003 ms | 1066 ms | — | — | 3 |
| — | openai/gpt-oss-20b warming | fireworks | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1031 ms | 2868 ms | 1031 ms | 2867 ms | — | — | 6 |
| — | qwen/qwen3-omni-30b-a3b-instruct warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1038 ms | 1038 ms | 1038 ms | 1038 ms | — | — | 1 |
| — | thinkingmachines/inkling-1m warming | baseten | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1045 ms | 3003 ms | 1044 ms | 3003 ms | 94 tok/s warming | — | 6 |
| — | google/gemma-4-31b-it warming | lightning | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1047 ms | 1047 ms | 1047 ms | 1047 ms | 102 tok/s warming | — | 1 |
| — | openai/gpt-5.6-terra warming | openai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1058 ms | 3245 ms | 1058 ms | 3245 ms | 25 tok/s warming | — | 3 |
| — | Qwen/Qwen3-30B-A3B-Instruct-2507 warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1066 ms | 2831 ms | 1066 ms | 2831 ms | — | — | 4 |
| — | deepseek/deepseek-r1-distill-llama-70b warming | novita | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1093 ms | 1093 ms | 1093 ms | 1093 ms | — | — | 1 |
| — | meta-llama/llama-3.2-3b-instruct warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1100 ms | 2950 ms | 1100 ms | 2950 ms | — | — | 2 |
| — | google/gemini-2.5-flash-lite warming | google-vertex | 100.00% | 100.00% | 100.00% 15s | 100.00% | 1102 ms | 3359 ms | 1102 ms | 3359 ms | 137 tok/s warming | — | 3 |
| — | mistralai/ministral-3-14b-instruct warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1112 ms | 3109 ms | 1112 ms | 3109 ms | — | — | 2 |
| — | openai/gpt-3.5-turbo warming | lightning | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1112 ms | 2422 ms | 1112 ms | 2421 ms | — | — | 4 |
| — | nvidia/nemotron-3-super-120b-a12b warming | crusoe | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1134 ms | 4319 ms | 1134 ms | 4319 ms | — | — | 3 |
| — | qwen/qwen3.6-plus warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1153 ms | 1153 ms | 1153 ms | 1153 ms | — | — | 1 |
| — | openai/gpt-oss-120b warming | fireworks | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1170 ms | 1335 ms | 1170 ms | 1335 ms | 116 tok/s warming | — | 5 |
| — | arcee-ai/trinity-large-thinking warming | parasail | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1180 ms | 1587 ms | 1180 ms | 1586 ms | — | — | 3 |
| — | qwen/qwen3-vl-flash-2025-10-15 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1185 ms | 1185 ms | 1185 ms | 1185 ms | — | — | 1 |
| — | tencent/hy3 warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1187 ms | 1427 ms | 1187 ms | 1427 ms | 40 tok/s warming | — | 2 |
| — | moonshotai/moonshot-v1-32k-vision-preview warming | kimi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1189 ms | 2978 ms | 1189 ms | 2978 ms | — | — | 4 |
| — | openai/gpt-oss-120b warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1240 ms | 2611 ms | 1240 ms | 2610 ms | 48 tok/s warming | — | 3 |
| — | moonshotai/kimi-k2.5 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1243 ms | 1243 ms | 1243 ms | 1243 ms | — | — | 1 |
| — | kwaipilot/kat-coder-pro-v2 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1257 ms | 1257 ms | 1257 ms | 1257 ms | — | — | 1 |
| — | bytedance/doubao-seed-2.0-pro-260215 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1258 ms | 3883 ms | 1258 ms | 3883 ms | — | — | 2 |
| — | qwen/qwen2.5-coder-32b-instruct warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1260 ms | 1436 ms | 1260 ms | 1436 ms | — | — | 2 |
| — | qwen/qwen3.6-35b-a3b warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1268 ms | 4174 ms | 1268 ms | 4174 ms | — | — | 2 |
| — | mistralai/ministral-3b-2512 warming | mistral | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1276 ms | 3887 ms | 1276 ms | 3887 ms | — | — | 8 |
| — | mistralai/codestral-2508 warming | mistral | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1282 ms | 2939 ms | 1282 ms | 2939 ms | — | — | 5 |
| — | thinkingmachines/inkling-small warming | baseten | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1285 ms | 2674 ms | 1285 ms | 2674 ms | 69 tok/s warming | — | 4 |
| — | google/gemini-3.1-flash-image-preview warming | google-ai-studio | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1334 ms | 3522 ms | 1334 ms | 3522 ms | — | — | 5 |
| — | deepseek/deepseek-v4-flash warming | morph | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1338 ms | 2415 ms | 1337 ms | 2415 ms | 42 tok/s warming | — | 4 |
| — | qwen/qwen3.5-397b-a17b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1338 ms | 1673 ms | 1338 ms | 1673 ms | — | — | 2 |
| — | z-ai/glm-5 warming | digitalocean | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1341 ms | 4260 ms | 1341 ms | 4260 ms | — | — | 3 |
| — | cerebras/gemma-4-31b warming | cerebras | 100.00% | 100.00% | 100.00% 15s | 100.00% | 1343 ms | 4594 ms | 1342 ms | 4594 ms | — | — | 8 |
| — | google/gemma-4-26b-a4b-it warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1347 ms | 7467 ms | 1347 ms | 7467 ms | — | — | 2 |
| — | NousResearch/Hermes-4-405B warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1349 ms | 2441 ms | 1349 ms | 2441 ms | — | — | 2 |
| — | NousResearch/Hermes-4-70B warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1351 ms | 1889 ms | 1351 ms | 1888 ms | — | — | 2 |
| — | qwen/qwen-plus-2025-12-01 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1351 ms | 2600 ms | 1351 ms | 2600 ms | — | — | 2 |
| — | openai/gpt-4.1-nano warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1355 ms | 4005 ms | 1355 ms | 4004 ms | — | — | 3 |
| — | qwen/qwen3.5-122b-a10b warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1375 ms | 3528 ms | 1375 ms | 3528 ms | — | — | 2 |
| — | qwen/qwen3-235b-a22b-thinking-2507 warming | chutes | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1379 ms | 2171 ms | 1379 ms | 2171 ms | — | — | 4 |
| — | yutori/n1.5 warming | crusoe | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1382 ms | 4245 ms | 1382 ms | 4245 ms | — | — | 6 |
| — | qwen/qwen3.7-flash-2026-07-15 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1387 ms | 1387 ms | 1386 ms | 1386 ms | 123 tok/s warming | — | 1 |
| — | google/gemma-3-27b-it warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1397 ms | 1397 ms | 1396 ms | 1396 ms | — | — | 1 |
| — | minimax/minimax-m3 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1410 ms | 2343 ms | 1409 ms | 2343 ms | 82 tok/s warming | — | 2 |
| — | meta-llama/llama-3.3-70b-instruct warming | crusoe | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1424 ms | 1563 ms | 1424 ms | 1563 ms | 42 tok/s warming | — | 2 |
| — | openai/gpt-oss-120b warming | cerebras | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1424 ms | 2783 ms | 1424 ms | 2783 ms | 163 tok/s warming | — | 7 |
| — | mistralai/mistral-nemo warming | mistral | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1427 ms | 2921 ms | 1427 ms | 2920 ms | — | — | 8 |
| — | meta-llama/llama-4-maverick warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1429 ms | 3670 ms | 1429 ms | 3670 ms | — | — | 4 |
| — | kwaipilot/kat-coder-pro-v2.5 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1442 ms | 1442 ms | 1441 ms | 1441 ms | — | — | 1 |
| — | qwen/qwen3.6-27b warming | venice | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1442 ms | 3444 ms | 1442 ms | 3444 ms | — | — | 7 |
| — | openai/gpt-5.2 warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1443 ms | 2857 ms | 1443 ms | 2857 ms | — | — | 3 |
| — | qwen/qwen3-30b-a3b-instruct-2507 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1446 ms | 1446 ms | 1446 ms | 1446 ms | — | — | 1 |
| — | qwen/qwen3-vl-flash warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1452 ms | 1703 ms | 1452 ms | 1703 ms | — | — | 2 |
| — | qwen/qwen3.5-27b warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1452 ms | 2152 ms | 1451 ms | 2152 ms | — | — | 3 |
| — | aisingapore/gemma-sea-lion-v4-27b-it warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1456 ms | 3286 ms | 1456 ms | 3286 ms | — | — | 5 |
| — | deepseek/deepseek-r1-distill-qwen-32b warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1459 ms | 2715 ms | 1459 ms | 2715 ms | — | — | 2 |
| — | meta-llama/llama-3.2-1b-instruct warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1459 ms | 2921 ms | 1459 ms | 2921 ms | — | — | 4 |
| — | qwen/qwq-32b warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1462 ms | 2407 ms | 1461 ms | 2407 ms | — | — | 3 |
| — | z-ai/glm-5.1 warming | friendli | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1466 ms | 4244 ms | 1466 ms | 4244 ms | — | — | 6 |
| — | google/gemini-2.5-flash warming | google-vertex | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1474 ms | 4042 ms | 1474 ms | 4042 ms | — | — | 9 |
| — | moonshotai/kimi-k2.6 warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1474 ms | 3997 ms | 1474 ms | 3997 ms | 29 tok/s warming | — | 6 |
| — | minimax/minimax-m3 warming | telnyx | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1475 ms | 3190 ms | 1475 ms | 3190 ms | 92 tok/s warming | — | 8 |
| — | ibm-granite/granite-4.0-h-micro warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1477 ms | 1665 ms | 1476 ms | 1665 ms | — | — | 2 |
| — | openai/gpt-oss-120b warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1488 ms | 3146 ms | 1488 ms | 3146 ms | 21 tok/s warming | — | 2 |
| — | cerebras/zai-glm-4.7 warming | cerebras | 100.00% | 100.00% | 100.00% 15s | 100.00% | 1491 ms | 6373 ms | 1491 ms | 6373 ms | 69 tok/s warming | — | 9 |
| — | nvidia/nemotron-3-nano-30b-a3b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1496 ms | 1496 ms | 1496 ms | 1496 ms | — | — | 1 |
| — | qwen/qwen3-vl-30b-a3b-instruct warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1496 ms | 1496 ms | 1496 ms | 1496 ms | — | — | 1 |
| — | nousresearch/hermes-3-llama-3.1-405b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1498 ms | 4582 ms | 1498 ms | 4582 ms | — | — | 2 |
| — | bytedance/ui-tars-1.5-7b warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1500 ms | 1500 ms | 1500 ms | 1500 ms | — | — | 1 |
| — | meta-llama/llama-4-scout-17b-16e-instruct warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1502 ms | 1573 ms | 1502 ms | 1573 ms | — | — | 3 |
| — | deepseek/deepseek-v3.2 warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1507 ms | 1507 ms | 1507 ms | 1507 ms | — | — | 1 |
| — | minimax/minimax-m2.7 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1510 ms | 1510 ms | 1509 ms | 1509 ms | — | — | 1 |
| — | openai/gpt-5.6-sol warming | openai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1512 ms | 2067 ms | 1511 ms | 2066 ms | 61 tok/s warming | — | 3 |
| — | moonshotai/Kimi-K2.7-Code warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1514 ms | 1831 ms | 1513 ms | 1831 ms | 72 tok/s warming | — | 3 |
| — | qwen/qwen3-max warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1519 ms | 1519 ms | 1519 ms | 1519 ms | — | — | 1 |
| — | meta-llama/Llama-3.3-70B-Instruct warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1521 ms | 15839 ms | 1521 ms | 15839 ms | — | — | 8 |
| — | z-ai/glm-5.1 warming | digitalocean | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1535 ms | 1554 ms | 1535 ms | 1553 ms | — | — | 2 |
| — | anthropic/claude-sonnet-4.5 warming | anthropic | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1543 ms | 3181 ms | 1543 ms | 3180 ms | — | — | 5 |
| — | bytedance/doubao-seed-character-260628 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1544 ms | 1544 ms | 1544 ms | 1544 ms | — | — | 1 |
| — | moonshotai/kimi-k2.6 warming | fireworks | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1551 ms | 3076 ms | 1551 ms | 3076 ms | 30 tok/s warming | — | 6 |
| — | moonshotai/kimi-k2.7-code-highspeed warming | kimi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1552 ms | 3437 ms | 1552 ms | 3437 ms | — | — | 6 |
| — | google/gemma-4-31b-it warming | friendli | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1558 ms | 17305 ms | 1558 ms | 17304 ms | 99 tok/s warming | — | 5 |
| — | MiniMaxAI/MiniMax-M3 warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1563 ms | 1563 ms | 1563 ms | 1563 ms | 85 tok/s warming | — | 1 |
| — | moonshotai/moonshot-v1-8k warming | kimi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1564 ms | 3241 ms | 1564 ms | 3241 ms | — | — | 6 |
| — | z-ai/glm-4.7 warming | venice | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1570 ms | 2103 ms | 1570 ms | 2102 ms | — | — | 5 |
| — | moonshotai/kimi-k2.5 warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1572 ms | 7628 ms | 1572 ms | 7628 ms | — | — | 5 |
| — | openai/gpt-4o warming | lightning | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1572 ms | 3263 ms | 1572 ms | 3262 ms | — | — | 5 |
| — | openai/gpt-5.6-luna warming | openai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1573 ms | 3336 ms | 1572 ms | 3336 ms | 67 tok/s warming | — | 3 |
| — | moonshotai/kimi-k2.7-code warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1585 ms | 3940 ms | 1585 ms | 3940 ms | 39 tok/s warming | — | 2 |
| — | nvidia/nemotron-nano-12b-v2-vl warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1586 ms | 2944 ms | 1586 ms | 2944 ms | — | — | 6 |
| — | moonshotai/moonshot-v1-128k warming | kimi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1587 ms | 2002 ms | 1587 ms | 2002 ms | — | — | 6 |
| — | moonshotai/kimi-k2.7-code warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1589 ms | 3547 ms | 1588 ms | 3547 ms | 52 tok/s warming | — | 3 |
| — | google/gemma-4-26b-a4b-it warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1590 ms | 1590 ms | 1590 ms | 1590 ms | — | — | 1 |
| — | openai/gpt-oss-120b warming | baseten | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1590 ms | 2820 ms | 1590 ms | 2819 ms | 121 tok/s warming | — | 3 |
| — | z-ai/glm-5.1 warming | telnyx | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1590 ms | 3844 ms | 1590 ms | 3844 ms | — | — | 2 |
| — | qwen/qwen3-14b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1595 ms | 5363 ms | 1594 ms | 5363 ms | — | — | 4 |
| — | deepseek/deepseek-v3.2 warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1596 ms | 3431 ms | 1596 ms | 3431 ms | — | — | 3 |
| — | openai/gpt-oss-120b warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1597 ms | 1597 ms | 1597 ms | 1597 ms | 58 tok/s warming | — | 1 |
| — | meituan-longcat/longcat-2.0 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1617 ms | 1712 ms | 1617 ms | 1712 ms | — | — | 3 |
| — | nvidia/Nemotron-3-Nano-Omni warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1618 ms | 1618 ms | 1617 ms | 1617 ms | — | — | 1 |
| — | deepseek/deepseek-v4-flash-0731 warming | fireworks | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1619 ms | 14793 ms | 1619 ms | 14793 ms | 57 tok/s warming | — | 6 |
| — | openai/gpt-oss-120b warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1619 ms | 1619 ms | 1619 ms | 1619 ms | 225 tok/s warming | — | 1 |
| — | moonshotai/kimi-k2.7-code warming | kimi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1623 ms | 3053 ms | 1623 ms | 3053 ms | 29 tok/s warming | — | 2 |
| — | mistralai/mistral-large warming | mistral | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1624 ms | 3025 ms | 1624 ms | 3024 ms | — | — | 8 |
| — | nvidia/nemotron-3-ultra-550b warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1637 ms | 3424 ms | 1637 ms | 3424 ms | — | — | 3 |
| — | google/gemini-2.5-flash warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1641 ms | 3800 ms | 1641 ms | 3800 ms | — | — | 4 |
| — | minimax/minimax-m2.7 warming | morph | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1642 ms | 2949 ms | 1642 ms | 2948 ms | — | — | 7 |
| — | google/gemini-3-flash-preview warming | google-vertex | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1643 ms | 3194 ms | 1643 ms | 3194 ms | — | — | 5 |
| — | google/gemma-3-27b-it warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1645 ms | 1645 ms | 1645 ms | 1645 ms | — | — | 1 |
| — | bytedance/doubao-seed-1.6-flash-250828 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1649 ms | 1649 ms | 1649 ms | 1649 ms | — | — | 1 |
| — | meta-llama/llama-3.1-70b-instruct warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1653 ms | 1653 ms | 1653 ms | 1653 ms | — | — | 1 |
| — | anthropic/claude-haiku-4.5 warming | anthropic | 100.00% | 100.00% | 100.00% 15s | 100.00% | 1657 ms | 4395 ms | 1657 ms | 4395 ms | — | — | 7 |
| — | qwen/qwen3-235b-a22b-fp8 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1658 ms | 3284 ms | 1658 ms | 3284 ms | — | — | 2 |
| — | mistralai/ministral-8b-2512 warming | mistral | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1661 ms | 4660 ms | 1661 ms | 4660 ms | — | — | 9 |
| — | qwen/qwen3.7-plus warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1667 ms | 3852 ms | 1667 ms | 3852 ms | — | — | 2 |
| — | z-ai/glm-4.7 warming | baseten | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1668 ms | 3298 ms | 1668 ms | 3298 ms | — | — | 7 |
| — | google/gemini-3.5-flash-lite warming | google-vertex | 100.00% | 100.00% | 100.00% 15s | 100.00% | 1678 ms | 3344 ms | 1678 ms | 3344 ms | 42 tok/s warming | — | 5 |
| — | z-ai/glm-5 warming | venice | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1678 ms | 13671 ms | 1677 ms | 13671 ms | — | — | 3 |
| — | meta-llama/llama-3.3-70b-instruct warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1689 ms | 5313 ms | 1689 ms | 5313 ms | — | — | 2 |
| — | meta-llama/llama-3.1-8b-instruct warming | telnyx | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1695 ms | 3177 ms | 1695 ms | 3177 ms | — | — | 7 |
| — | moonshotai/kimi-k2.7-code warming | together | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1696 ms | 3264 ms | 1696 ms | 3264 ms | — | — | 5 |
| — | moonshotai/kimi-k3 warming | together | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1707 ms | 5653 ms | 1707 ms | 5653 ms | 31 tok/s warming | — | 3 |
| — | qwen/qwen3.5-35b-a3b warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1717 ms | 1717 ms | 1716 ms | 1716 ms | — | — | 1 |
| — | z-ai/glm-4.6 warming | venice | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1724 ms | 3483 ms | 1724 ms | 3483 ms | — | — | 4 |
| — | z-ai/glm-5.2 warming | siliconflow | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1729 ms | 3646 ms | 1729 ms | 3646 ms | 30 tok/s warming | — | 3 |
| — | z-ai/glm-4.6v warming | zai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1731 ms | 4304 ms | 1731 ms | 4304 ms | — | — | 7 |
| — | tencent/hunyuan-a13b-instruct warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1745 ms | 2017 ms | 1745 ms | 2017 ms | — | — | 2 |
| — | anthropic/claude-sonnet-5 warming | anthropic | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1755 ms | 3479 ms | 1755 ms | 3479 ms | — | — | 6 |
| — | google/gemma-4-26b-a4b-it warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1756 ms | 1756 ms | 1756 ms | 1756 ms | — | — | 1 |
| — | moonshotai/Kimi-K2.6 warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1759 ms | 3246 ms | 1759 ms | 3246 ms | — | — | 4 |
| — | openai/o3-mini warming | openai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1760 ms | 4429 ms | 1760 ms | 4429 ms | — | — | 2 |
| — | google/gemini-3.6-flash warming | google-ai-studio | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1767 ms | 3482 ms | 1767 ms | 3482 ms | — | — | 4 |
| — | nvidia/nemotron-3-nano-omni-reasoning-30b-a3b warming | crusoe | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1768 ms | 5749 ms | 1768 ms | 5749 ms | — | — | 2 |
| — | moonshotai/kimi-k2.6 warming | wafer | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1772 ms | 1772 ms | 1771 ms | 1771 ms | 27 tok/s warming | — | 1 |
| — | qwen/qwen3-235b-a22b-instruct-2507 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1786 ms | 1786 ms | 1786 ms | 1786 ms | — | — | 1 |
| — | meta-llama/llama-3.3-70b-instruct warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1789 ms | 4506 ms | 1789 ms | 4506 ms | 106 tok/s warming | — | 3 |
| — | qwen/qwen3-coder-flash warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1800 ms | 4890 ms | 1800 ms | 4890 ms | — | — | 5 |
| — | qwen/qwen3.5-flash warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1802 ms | 1802 ms | 1802 ms | 1802 ms | — | — | 1 |
| — | deepseek/deepseek-v4-flash warming | crusoe | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1812 ms | 4051 ms | 1812 ms | 4051 ms | 82 tok/s warming | — | 7 |
| — | qwen/qwen3.5-35b-a3b warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1815 ms | 1815 ms | 1815 ms | 1815 ms | — | — | 1 |
| — | qwen/qwen3.5-35b-a3b warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1815 ms | 2581 ms | 1814 ms | 2581 ms | — | — | 2 |
| — | qwen/qwen3.5-9b warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1821 ms | 3558 ms | 1820 ms | 3558 ms | — | — | 3 |
| — | z-ai/glm-5v-turbo warming | zai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1831 ms | 3588 ms | 1831 ms | 3588 ms | — | — | 8 |
| — | z-ai/glm-4.6 warming | zai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1832 ms | 4483 ms | 1832 ms | 4483 ms | — | — | 3 |
| — | google/gemma-3-12b-it warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1835 ms | 2790 ms | 1835 ms | 2790 ms | — | — | 3 |
| — | openai/gpt-oss-20b warming | together | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1854 ms | 3375 ms | 1854 ms | 3375 ms | — | — | 5 |
| — | Qwen/Qwen2.5-VL-72B-Instruct warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1855 ms | 1855 ms | 1854 ms | 1854 ms | — | — | 1 |
| — | mistralai/mistral-nemo warming | chutes | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1859 ms | 3570 ms | 1859 ms | 3570 ms | — | — | 6 |
| — | z-ai/glm-5.2 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1860 ms | 3024 ms | 1860 ms | 3024 ms | 46 tok/s warming | — | 2 |
| — | z-ai/glm-4.7-flash warming | venice | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1867 ms | 18598 ms | 1867 ms | 18597 ms | — | — | 6 |
| — | qwen/qwen-2.5-7b-instruct warming | together | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1868 ms | 4057 ms | 1868 ms | 4057 ms | — | — | 3 |
| — | google/gemma-4-31b-it warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1870 ms | 5304 ms | 1870 ms | 5304 ms | 37 tok/s warming | — | 3 |
| — | openai/gpt-4.1 warming | lightning | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1870 ms | 3544 ms | 1870 ms | 3543 ms | — | — | 7 |
| — | z-ai/glm-4.7 warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1874 ms | 1874 ms | 1873 ms | 1873 ms | — | — | 1 |
| — | google/gemini-3.5-flash warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1882 ms | 2095 ms | 1880 ms | 2094 ms | 57 tok/s warming | — | 2 |
| — | qwen/qwen3.5-27b warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1882 ms | 1882 ms | 1881 ms | 1881 ms | — | — | 1 |
| — | google/gemini-3.1-flash-lite-preview warming | lightning | 100.00% | 100.00% | 100.00% 15s | 100.00% | 1906 ms | 3425 ms | 1906 ms | 3425 ms | — | — | 4 |
| — | openai/gpt-oss-20b warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1913 ms | 2733 ms | 1912 ms | 2733 ms | — | — | 2 |
| — | z-ai/glm-5-turbo warming | zai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1927 ms | 24673 ms | 1927 ms | 24673 ms | — | — | 6 |
| — | deepseek/deepseek-v4-pro warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1928 ms | 2217 ms | 1928 ms | 2217 ms | 104 tok/s warming | — | 3 |
| — | google/gemma-3-4b-it warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1944 ms | 3238 ms | 1943 ms | 3238 ms | — | — | 2 |
| — | moonshotai/kimi-k2.5 warming | kimi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1945 ms | 4050 ms | 1945 ms | 4049 ms | — | — | 8 |
| — | minimax/minimax-m2.7 warming | minimax | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1951 ms | 4094 ms | 1951 ms | 4093 ms | — | — | 9 |
| — | anthropic/claude-opus-4.7 warming | anthropic | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1962 ms | 3989 ms | 1962 ms | 3988 ms | — | — | 7 |
| — | moonshotai/moonshot-v1-128k-vision-preview warming | kimi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1962 ms | 3133 ms | 1961 ms | 3132 ms | — | — | 5 |
| — | z-ai/glm-5.1 warming | venice | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1962 ms | 6654 ms | 1962 ms | 6654 ms | — | — | 3 |
| — | thinkingmachines/inkling warming | together | 100.00% | 100.00% | 100.00% 45s | 100.00% | 1987 ms | 3882 ms | 1987 ms | 3882 ms | 24 tok/s warming | — | 2 |
| — | qwen/qwen3.5-flash-2026-02-23 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 1992 ms | 1992 ms | 1991 ms | 1991 ms | — | — | 1 |
| — | z-ai/glm-5.2 warming | morph | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2001 ms | 14247 ms | 2001 ms | 14247 ms | 53 tok/s warming | — | 9 |
| — | deepseek/deepseek-v4-pro warming | crusoe | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2009 ms | 4265 ms | 2009 ms | 4265 ms | 61 tok/s warming | — | 8 |
| — | qwen/qwen3.5-122b-a10b warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2018 ms | 2018 ms | 2017 ms | 2017 ms | — | — | 1 |
| — | moonshotai/moonshot-v1-8k-vision-preview warming | kimi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2020 ms | 3509 ms | 2020 ms | 3509 ms | — | — | 5 |
| — | z-ai/glm-4.5v warming | zai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2021 ms | 3686 ms | 2021 ms | 3686 ms | — | — | 3 |
| — | moonshotai/kimi-k2.6 warming | chutes | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2025 ms | 3640 ms | 2025 ms | 3640 ms | 34 tok/s warming | — | 6 |
| — | z-ai/glm-5.1 warming | chutes | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2027 ms | 4202 ms | 2027 ms | 4201 ms | — | — | 4 |
| — | z-ai/glm-4.5-air warming | zai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2035 ms | 4062 ms | 2035 ms | 4062 ms | — | — | 4 |
| — | z-ai/glm-5.1 warming | siliconflow | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2040 ms | 4618 ms | 2039 ms | 4618 ms | — | — | 2 |
| — | moonshotai/kimi-k2.6 warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2045 ms | 2045 ms | 2044 ms | 2044 ms | — | — | 1 |
| — | nvidia/nemotron-3-super-120b-a12b warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2047 ms | 3718 ms | 2046 ms | 3717 ms | — | — | 2 |
| — | deepseek/deepseek-v4-pro warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2054 ms | 2054 ms | 2054 ms | 2054 ms | — | — | 1 |
| — | google/gemini-2.5-flash-lite warming | atlas-cloud | 100.00% | 100.00% | 100.00% 15s | 100.00% | 2056 ms | 2056 ms | 2056 ms | 2056 ms | 150 tok/s warming | — | 1 |
| — | google/gemma-4-e2b-it warming | neurometric | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2063 ms | 3307 ms | 2063 ms | 3307 ms | — | — | 8 |
| — | anthropic/claude-opus-4.8 warming | anthropic | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2066 ms | 7770 ms | 2066 ms | 7770 ms | 101 tok/s warming | — | 6 |
| — | zai-org/GLM-5.2 warming | nebius | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2073 ms | 2073 ms | 2073 ms | 2073 ms | 101 tok/s warming | — | 1 |
| — | nvidia/nemotron-3-nano-30b-a3b warming | crusoe | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2074 ms | 4626 ms | 2073 ms | 4625 ms | — | — | 5 |
| — | deepseek/deepseek-v4-pro warming | baseten | 100.00% | 100.00% | 85.71% 20s | 100.00% | 2080 ms | 22072 ms | 2080 ms | 22072 ms | 72 tok/s warming | — | 7 |
| — | qwen/qwen3-vl-plus-2025-09-23 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2085 ms | 2595 ms | 2085 ms | 2595 ms | — | — | 2 |
| — | qwen/qwen-mt-flash warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2087 ms | 2087 ms | 2087 ms | 2087 ms | — | — | 1 |
| — | nvidia/nemotron-3-nano-30b-a3b warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2124 ms | 2124 ms | 2124 ms | 2124 ms | — | — | 1 |
| — | qwen/qwen3-omni-30b-a3b-thinking warming | novita | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2129 ms | 2129 ms | 2129 ms | 2129 ms | — | — | 1 |
| — | bytedance/doubao-seed-2.1-pro-260628 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2132 ms | 2132 ms | 2132 ms | 2132 ms | — | — | 1 |
| — | meta-llama/llama-3.1-8b-instruct-fp8 warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2145 ms | 3231 ms | 2145 ms | 3230 ms | — | — | 3 |
| — | openai/gpt-5.6-terra warming | atlas-cloud | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2145 ms | 2145 ms | 2145 ms | 2145 ms | 33 tok/s warming | — | 1 |
| — | z-ai/glm-5.2 warming | makora | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2148 ms | 4159 ms | 2147 ms | 4159 ms | — | — | 7 |
| — | z-ai/glm-4.5-air warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2151 ms | 2151 ms | 2150 ms | 2150 ms | — | 1 provider_auth_config |
1 |
| — | lgai-exaone/k-exaone-236b-a23b warming | friendli | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2156 ms | 2959 ms | 2156 ms | 2959 ms | — | — | 7 |
| — | qwen/qwen3-vl-8b-thinking warming | alibaba | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2159 ms | 2243 ms | 2159 ms | 2243 ms | — | — | 2 |
| — | qwen/qwen3.7-max-2026-06-08 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2162 ms | 2162 ms | 2162 ms | 2162 ms | — | — | 1 |
| — | google/gemini-3.5-flash warming | google-vertex | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2163 ms | 3384 ms | 2163 ms | 3384 ms | 60 tok/s warming | — | 7 |
| — | qwen/qwen3.5-plus warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2175 ms | 2175 ms | 2174 ms | 2174 ms | — | — | 1 |
| — | bytedance/doubao-seed-1.8-251228 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2188 ms | 2188 ms | 2187 ms | 2187 ms | — | — | 1 |
| — | qwen/qwen3-max-preview warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2197 ms | 2197 ms | 2197 ms | 2197 ms | — | — | 1 |
| — | z-ai/glm-5.1 warming | crusoe | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2205 ms | 5014 ms | 2205 ms | 5014 ms | — | — | 4 |
| — | deepseek/deepseek-r1-0528 warming | novita | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2206 ms | 3972 ms | 2206 ms | 3972 ms | — | — | 2 |
| — | deepseek/deepseek-v4-flash warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2209 ms | 2209 ms | 2209 ms | 2209 ms | 32 tok/s warming | — | 1 |
| — | qwen/qwen3.6-27b warming | morph | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2211 ms | 3936 ms | 2210 ms | 3936 ms | — | — | 7 |
| — | z-ai/glm-5.2 warming | zai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2222 ms | 3949 ms | 2222 ms | 3949 ms | 67 tok/s warming | — | 3 |
| — | z-ai/glm-5 warming | zai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2243 ms | 4186 ms | 2243 ms | 4185 ms | — | — | 5 |
| — | qwen/qwen3.6-plus warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2253 ms | 3482 ms | 2253 ms | 3482 ms | — | — | 3 |
| — | minimax/minimax-m2.5 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2262 ms | 2262 ms | 2262 ms | 2262 ms | — | — | 1 |
| — | deepseek/deepseek-v3.2-exp warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2263 ms | 4697 ms | 2262 ms | 4696 ms | — | — | 2 |
| — | z-ai/glm-5.2 warming | together | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2263 ms | 18324 ms | 2263 ms | 18324 ms | — | — | 7 |
| — | moonshotai/kimi-k3 warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2269 ms | 5087 ms | 2269 ms | 5086 ms | 25 tok/s warming | — | 2 |
| — | moonshotai/kimi-k2.5 warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2274 ms | 4701 ms | 2274 ms | 4701 ms | — | — | 4 |
| — | qwen/qwen3-next-80b-a3b-instruct warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2274 ms | 2274 ms | 2274 ms | 2274 ms | — | — | 1 |
| — | minimax/minimax-m2.5-highspeed warming | minimax | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2291 ms | 4164 ms | 2291 ms | 4164 ms | — | — | 4 |
| — | anthropic/claude-opus-5 warming | anthropic | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2296 ms | 4510 ms | 2296 ms | 4510 ms | 60 tok/s warming | — | 4 |
| — | deepseek/deepseek-v3.1-terminus warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2306 ms | 3622 ms | 2305 ms | 3622 ms | — | — | 4 |
| — | nvidia/nemotron-3-super-120b warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2307 ms | 12257 ms | 2307 ms | 12257 ms | — | — | 3 |
| — | openai/gpt-5.6-sol warming | atlas-cloud | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2329 ms | 2329 ms | 2329 ms | 2329 ms | 45 tok/s warming | — | 1 |
| — | moonshotai/kimi-k2.5 warming | telnyx | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2335 ms | 3886 ms | 2335 ms | 3886 ms | — | 2 probe_config_error |
5 |
| — | moonshotai/kimi-k3 warming | telnyx | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2340 ms | 4150 ms | 2340 ms | 4150 ms | 39 tok/s warming | — | 8 |
| — | google/gemma-3n-e4b-it warming | together | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2342 ms | 5021 ms | 2342 ms | 5021 ms | — | — | 2 |
| — | nvidia/Cosmos3-Super-Reasoner warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2353 ms | 2995 ms | 2353 ms | 2995 ms | 51 tok/s warming | — | 4 |
| — | z-ai/glm-5.2-nvfp4 warming | makora | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2378 ms | 3743 ms | 2378 ms | 3743 ms | — | — | 9 |
| — | minimax/minimax-m3 warming | morph | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2384 ms | 4740 ms | 2384 ms | 4740 ms | 42 tok/s warming | — | 7 |
| — | qwen/qwen3.6-35b-a3b warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2392 ms | 2392 ms | 2392 ms | 2392 ms | — | — | 1 |
| — | moonshotai/kimi-k2.6 warming | telnyx | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2393 ms | 3292 ms | 2392 ms | 3292 ms | — | — | 2 |
| — | minimax/minimax-m2.7 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2401 ms | 2401 ms | 2401 ms | 2401 ms | — | — | 1 |
| — | meta-llama/llama-3.3-70b-instruct warming | together | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2412 ms | 2632 ms | 2412 ms | 2632 ms | 68 tok/s warming | — | 2 |
| — | openbmb/MiniCPM-V-4_5 warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2429 ms | 2696 ms | 2428 ms | 2696 ms | 85 tok/s warming | — | 3 |
| — | qwen/qwen3-vl-235b-a22b-thinking warming | alibaba | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2454 ms | 2454 ms | 2454 ms | 2454 ms | — | — | 1 |
| — | google/gemma-3-27b-it warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2462 ms | 10163 ms | 2462 ms | 10162 ms | — | — | 4 |
| — | qwen/qwen-2.5-72b-instruct warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2480 ms | 7456 ms | 2480 ms | 7456 ms | — | — | 2 |
| — | z-ai/glm-5-turbo warming | venice | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2481 ms | 4009 ms | 2480 ms | 4009 ms | — | — | 4 |
| — | qwen/qwen3.5-plus warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2484 ms | 3565 ms | 2484 ms | 3564 ms | — | — | 2 |
| — | qwen/qwen3.5-27b warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2488 ms | 11999 ms | 2488 ms | 11998 ms | — | — | 2 |
| — | qwen/qwen3-coder-plus-2025-07-22 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2494 ms | 2494 ms | 2494 ms | 2494 ms | — | — | 1 |
| — | anthropic/claude-opus-4.5 warming | anthropic | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2498 ms | 7409 ms | 2498 ms | 7409 ms | — | — | 6 |
| — | deepseek/deepseek-v3.2 warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2517 ms | 3994 ms | 2516 ms | 3993 ms | — | 2 unsupported_route |
3 |
| — | qwen/qwen3-vl-235b-a22b-thinking warming | novita | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2523 ms | 2523 ms | 2523 ms | 2523 ms | — | — | 1 |
| — | google/gemini-3.5-flash warming | lightning | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2530 ms | 4284 ms | 2529 ms | 4284 ms | 40 tok/s warming | — | 4 |
| — | openai/gpt-5-nano warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2555 ms | 4835 ms | 2554 ms | 4835 ms | — | — | 5 |
| — | nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2580 ms | 2723 ms | 2580 ms | 2722 ms | — | — | 2 |
| — | google/gemini-3.1-flash-lite warming | google-vertex | 100.00% | 100.00% | 100.00% 15s | 100.00% | 2581 ms | 3877 ms | 2580 ms | 3877 ms | — | — | 7 |
| — | google/gemini-2.5-flash warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2584 ms | 2584 ms | 2583 ms | 2583 ms | — | — | 1 |
| — | z-ai/glm-5.2 warming | crusoe | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2584 ms | 4984 ms | 2584 ms | 4983 ms | 154 tok/s warming | — | 5 |
| — | qwen/qwen3.5-397b-a17b warming | venice | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2587 ms | 4357 ms | 2587 ms | 4357 ms | — | — | 7 |
| — | deepseek/deepseek-v3.1-terminus warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2596 ms | 2596 ms | 2595 ms | 2595 ms | — | 1 unsupported_route |
1 |
| — | moonshotai/kimi-k2.5 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2599 ms | 2599 ms | 2599 ms | 2599 ms | — | — | 1 |
| — | qwen/qwen3.6-27b warming | chutes | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2600 ms | 2600 ms | 2599 ms | 2599 ms | — | — | 1 |
| — | qwen/qwen3-vl-32b-thinking warming | alibaba | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2623 ms | 2623 ms | 2623 ms | 2623 ms | — | — | 1 |
| — | qwen/qwen3-next-80b-a3b-thinking warming | alibaba | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2640 ms | 3001 ms | 2640 ms | 3001 ms | — | — | 2 |
| — | z-ai/glm-4.6 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2660 ms | 2660 ms | 2660 ms | 2660 ms | — | — | 1 |
| — | moonshotai/kimi-k2.6 warming | crusoe | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2678 ms | 2678 ms | 2678 ms | 2678 ms | — | — | 1 |
| — | google/gemma-4-31b-it warming | crusoe | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2695 ms | 3983 ms | 2695 ms | 3983 ms | 48 tok/s warming | — | 3 |
| — | meta-llama/llama-4-maverick-17b-128e-instruct-fp8 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2714 ms | 2714 ms | 2714 ms | 2714 ms | — | — | 1 |
| — | deepseek/deepseek-v3-0324 warming | crusoe | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2725 ms | 4292 ms | 2725 ms | 4292 ms | — | — | 5 |
| — | openai/gpt-4-turbo warming | lightning | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2733 ms | 3336 ms | 2732 ms | 3336 ms | — | — | 3 |
| — | Qwen/Qwen3-235B-A22B-Instruct-2507 warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2736 ms | 2886 ms | 2735 ms | 2886 ms | — | — | 3 |
| — | google/gemma-4-26b-a4b-it warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2746 ms | 3923 ms | 2746 ms | 3923 ms | — | — | 5 |
| — | google/gemini-2.5-flash warming | google-ai-studio | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2748 ms | 3595 ms | 2748 ms | 3595 ms | — | — | 8 |
| — | deepseek/deepseek-v3.1-terminus warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2783 ms | 2783 ms | 2782 ms | 2782 ms | — | — | 1 |
| — | meta-llama/llama-3.3-70b-instruct warming | telnyx | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2791 ms | 3158 ms | 2790 ms | 3157 ms | 30 tok/s warming | — | 8 |
| — | openai/gpt-oss-20b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2819 ms | 6416 ms | 2819 ms | 6416 ms | — | — | 3 |
| — | moonshotai/moonshot-v1-32k warming | kimi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2823 ms | 3785 ms | 2823 ms | 3785 ms | — | — | 4 |
| — | z-ai/glm-5 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2826 ms | 2826 ms | 2826 ms | 2826 ms | — | — | 1 |
| — | anthropic/claude-opus-4.6 warming | anthropic | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2831 ms | 7557 ms | 2830 ms | 7557 ms | — | — | 5 |
| — | mistralai/mistral-small-3.1-24b-instruct warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2862 ms | 3376 ms | 2862 ms | 3376 ms | — | — | 5 |
| — | deepseek/deepseek-v3.2 warming | friendli | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2869 ms | 7092 ms | 2869 ms | 7091 ms | — | — | 7 |
| — | z-ai/glm-5.2 warming | fireworks | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2869 ms | 4542 ms | 2868 ms | 4541 ms | 43 tok/s warming | — | 3 |
| — | google/gemini-3.1-flash-lite warming | deepinfra | 100.00% | 100.00% | 100.00% 15s | 100.00% | 2872 ms | 4578 ms | 2872 ms | 4578 ms | — | — | 3 |
| — | qwen/qwen2.5-vl-72b-instruct warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2882 ms | 3839 ms | 2882 ms | 3839 ms | — | — | 5 |
| — | openai/gpt-5 warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2894 ms | 3204 ms | 2894 ms | 3203 ms | — | — | 3 |
| — | google/gemini-2.5-pro warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2908 ms | 2908 ms | 2908 ms | 2908 ms | — | — | 1 |
| — | z-ai/glm-5.2 warming | baseten | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2912 ms | 3474 ms | 2912 ms | 3474 ms | 45 tok/s warming | — | 5 |
| — | z-ai/glm-5.2 warming | telnyx | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2950 ms | 3647 ms | 2950 ms | 3647 ms | 100 tok/s warming | — | 5 |
| — | deepseek/deepseek-v4-flash warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2953 ms | 2953 ms | 2952 ms | 2952 ms | 67 tok/s warming | — | 1 |
| — | moonshotai/kimi-k2.5 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2959 ms | 3334 ms | 2959 ms | 3334 ms | — | — | 2 |
| — | openai/gpt-5.5 warming | openai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2959 ms | 4125 ms | 2959 ms | 4124 ms | 53 tok/s warming | — | 5 |
| — | thedrummer/cydonia-24b-v4.1 warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2972 ms | 2978 ms | 2972 ms | 2977 ms | — | — | 2 |
| — | z-ai/glm-5.1 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 45s | 100.00% | 2975 ms | 2975 ms | 2975 ms | 2975 ms | — | — | 1 |
| — | anthropic/claude-sonnet-4.6 warming | anthropic | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2977 ms | 3480 ms | 2977 ms | 3479 ms | 66 tok/s warming | — | 5 |
| — | qwen/qwen3.6-35b-a3b warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 2999 ms | 3700 ms | 2998 ms | 3700 ms | — | — | 2 |
| — | tencent/hy3 warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3000 ms | 3000 ms | 3000 ms | 3000 ms | 57 tok/s warming | — | 1 |
| — | moonshotai/kimi-k2.6 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3003 ms | 3204 ms | 3003 ms | 3204 ms | — | — | 2 |
| — | meta-llama/llama-3.1-70b-instruct warming | telnyx | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3008 ms | 3008 ms | 3008 ms | 3008 ms | — | — | 1 |
| — | deepseek/deepseek-r1-distill-llama-70b warming | digitalocean | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3010 ms | 3031 ms | 3010 ms | 3031 ms | — | — | 2 |
| — | openai/gpt-3.5-turbo warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3011 ms | 3593 ms | 3011 ms | 3593 ms | — | — | 5 |
| — | gryphe/mythomax-l2-13b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3012 ms | 3012 ms | 3012 ms | 3012 ms | — | — | 1 |
| — | z-ai/glm-5.1 warming | zai | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3024 ms | 4939 ms | 3024 ms | 4939 ms | — | — | 3 |
| — | xiaomi/mimo-v2.5-pro warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3039 ms | 3039 ms | 3039 ms | 3039 ms | 31 tok/s warming | — | 1 |
| — | deepseek-ai/DeepSeek-V4-Pro warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3045 ms | 3489 ms | 3045 ms | 3489 ms | — | — | 2 |
| — | openai/gpt-5.4-mini warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3051 ms | 3318 ms | 3051 ms | 3317 ms | — | — | 3 |
| — | minimax/minimax-m3 warming | minimax | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3084 ms | 4577 ms | 3084 ms | 4577 ms | 69 tok/s warming | — | 9 |
| — | nvidia/nemotron-3-ultra-550b-a55b warming | baseten | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3087 ms | 3197 ms | 3087 ms | 3197 ms | 234 tok/s warming | — | 4 |
| — | google/gemini-3.1-pro-preview warming | google-vertex | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3097 ms | 3506 ms | 3097 ms | 3505 ms | 56 tok/s warming | — | 3 |
| — | qwen/qwen3-235b-a22b-thinking-2507 warming | venice | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3100 ms | 3462 ms | 3100 ms | 3461 ms | — | — | 4 |
| — | qwen/qwen3.5-9b warming | venice | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3101 ms | 3487 ms | 3100 ms | 3487 ms | — | — | 5 |
| — | google/gemini-3.6-flash warming | google-vertex | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3110 ms | 3648 ms | 3110 ms | 3648 ms | 45 tok/s warming | — | 3 |
| — | z-ai/glm-5.2-fast warming | baseten | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3111 ms | 3280 ms | 3111 ms | 3279 ms | 64 tok/s warming | — | 5 |
| — | moonshotai/kimi-k3 warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3184 ms | 3184 ms | 3184 ms | 3184 ms | 80 tok/s warming | — | 1 |
| — | qwen/qwen3.6-35b-a3b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3189 ms | 3189 ms | 3189 ms | 3189 ms | — | — | 1 |
| — | google/gemini-2.5-pro warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3201 ms | 4535 ms | 3200 ms | 4535 ms | — | — | 2 |
| — | google/gemma-4-31b-it warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3203 ms | 3523 ms | 3202 ms | 3523 ms | 29 tok/s warming | — | 2 |
| — | tencent/hy3 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3205 ms | 3205 ms | 3205 ms | 3205 ms | 85 tok/s warming | — | 1 |
| — | zai-org/autoglm-phone-9b-multilingual warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3210 ms | 3210 ms | 3210 ms | 3210 ms | — | — | 1 |
| — | moonshotai/kimi-k2.7-code warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3212 ms | 3987 ms | 3211 ms | 3987 ms | 65 tok/s warming | — | 2 |
| — | z-ai/glm-5.2 warming | chutes | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3219 ms | 9240 ms | 3219 ms | 9240 ms | 31 tok/s warming | — | 4 |
| — | minimax/minimax-m2.5 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3221 ms | 3221 ms | 3221 ms | 3221 ms | — | — | 1 |
| — | bytedance/doubao-seed-2.0-mini-260428 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3231 ms | 3231 ms | 3231 ms | 3231 ms | — | — | 1 |
| — | deepseek/deepseek-v3.1 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3248 ms | 3248 ms | 3248 ms | 3248 ms | — | — | 1 |
| — | google/gemini-3.1-pro-preview warming | lightning | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3257 ms | 6080 ms | 3257 ms | 6080 ms | 34 tok/s warming | — | 3 |
| — | openai/gpt-4 warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3272 ms | 3272 ms | 3272 ms | 3272 ms | — | — | 1 |
| — | google/gemini-3.5-flash-lite warming | google-ai-studio | 100.00% | 100.00% | 100.00% 15s | 100.00% | 3295 ms | 3447 ms | 3294 ms | 3447 ms | — | — | 5 |
| — | qwen/qwen3-235b-a22b-instruct-2507 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3299 ms | 3299 ms | 3299 ms | 3299 ms | — | — | 1 |
| — | Qwen/Qwen3-Next-80B-A3B-Thinking warming | nebius | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3330 ms | 4641 ms | 3330 ms | 4641 ms | — | — | 3 |
| — | deepseek/deepseek-v3.1-terminus warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3346 ms | 7381 ms | 3345 ms | 7380 ms | — | — | 3 |
| — | moonshotai/kimi-k2.6 warming | kimi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3357 ms | 7404 ms | 3357 ms | 7404 ms | 13 tok/s warming | — | 7 |
| — | inclusionai/ring-2.6-1t warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3358 ms | 3659 ms | 3357 ms | 3659 ms | — | — | 3 |
| — | deepseek/deepseek-v4-flash warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3404 ms | 3404 ms | 3404 ms | 3404 ms | 82 tok/s warming | — | 1 |
| — | Qwen/Qwen3-32B warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3411 ms | 3411 ms | 3411 ms | 3411 ms | — | — | 1 |
| — | moonshotai/kimi-k2.7-code warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3421 ms | 3421 ms | 3421 ms | 3421 ms | — | — | 1 |
| — | deepseek/deepseek-v4-flash warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3423 ms | 5163 ms | 3423 ms | 5162 ms | 20 tok/s warming | — | 3 |
| — | minimax/minimax-m2.5 warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3454 ms | 3454 ms | 3454 ms | 3454 ms | — | — | 1 |
| — | zai-org/glm-4.6 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3467 ms | 3467 ms | 3467 ms | 3467 ms | — | — | 1 |
| — | openai/gpt-4 warming | lightning | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3474 ms | 4188 ms | 3474 ms | 4187 ms | — | — | 3 |
| — | qwen/qwen-mt-turbo warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3492 ms | 3492 ms | 3492 ms | 3492 ms | — | — | 1 |
| — | z-ai/glm-5.2 warming | venice | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3492 ms | 17844 ms | 3492 ms | 17844 ms | 33 tok/s warming | — | 5 |
| — | minimaxai/minimax-m1-80k warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3494 ms | 3494 ms | 3494 ms | 3494 ms | — | — | 1 |
| — | minimax/minimax-m2.5-highspeed warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3508 ms | 3508 ms | 3508 ms | 3508 ms | — | — | 1 |
| — | moonshotai/kimi-k3 warming | kimi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3516 ms | 6527 ms | 3515 ms | 6527 ms | 26 tok/s warming | — | 5 |
| — | qwen/qwen-plus-2025-09-11 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3520 ms | 3520 ms | 3520 ms | 3520 ms | — | — | 1 |
| — | qwen/qwen3.6-27b warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3547 ms | 3547 ms | 3547 ms | 3547 ms | — | — | 1 |
| — | meta-llama/llama-4-maverick warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3548 ms | 3548 ms | 3548 ms | 3548 ms | — | — | 1 |
| — | nvidia/nemotron-3-ultra-550b-a55b warming | nebius | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3558 ms | 3558 ms | 3558 ms | 3558 ms | 112 tok/s warming | — | 1 |
| — | qwen/qwen3.5-plus-2026-02-15 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3561 ms | 4123 ms | 3560 ms | 4123 ms | — | — | 3 |
| — | qwen/qwen3-32b warming | chutes | 100.00% | 100.00% | 75.00% 20s | 100.00% | 3575 ms | 32652 ms | 3575 ms | 32651 ms | — | — | 8 |
| — | qwen/qwen3.7-plus warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3597 ms | 3597 ms | 3596 ms | 3596 ms | — | — | 1 |
| — | thinkingmachines/inkling-small warming | deepinfra | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3641 ms | 3641 ms | 3641 ms | 3641 ms | 63 tok/s warming | — | 1 |
| — | z-ai/glm-5v-turbo warming | siliconflow | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3646 ms | 4237 ms | 3646 ms | 4237 ms | — | — | 4 |
| — | google/gemini-3-flash-preview warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3680 ms | 3680 ms | 3680 ms | 3680 ms | — | — | 1 |
| — | baidu/ernie-4.5-vl-424b-a47b warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3713 ms | 3713 ms | 3712 ms | 3712 ms | — | — | 1 |
| — | qwen/qwen3.5-27b warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3723 ms | 3723 ms | 3723 ms | 3723 ms | — | — | 1 |
| — | zai-org/glm-4.6v warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3740 ms | 3740 ms | 3740 ms | 3740 ms | — | — | 1 |
| — | google/gemini-3-flash-preview warming | lightning | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3742 ms | 4783 ms | 3741 ms | 4783 ms | — | — | 3 |
| — | z-ai/glm-5 warming | deepinfra | 100.00% | 100.00% | 100.00% 45s | 100.00% | 3752 ms | 3752 ms | 3752 ms | 3752 ms | — | — | 1 |
| — | kwaipilot/kat-coder-air-v2.5 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3762 ms | 3762 ms | 3762 ms | 3762 ms | — | — | 1 |
| — | qwen/qwen3-coder-flash warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3776 ms | 3776 ms | 3775 ms | 3775 ms | — | — | 1 |
| — | microsoft/wizardlm-2-8x22b warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3797 ms | 3877 ms | 3796 ms | 3876 ms | — | — | 2 |
| — | openai/gpt-oss-120b warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3805 ms | 3805 ms | 3805 ms | 3805 ms | 28 tok/s warming | — | 1 |
| — | google/gemini-2.5-pro warming | lightning | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3821 ms | 4113 ms | 3821 ms | 4113 ms | — | — | 3 |
| — | minimax/minimax-m3 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3856 ms | 3856 ms | 3856 ms | 3856 ms | 83 tok/s warming | — | 1 |
| — | inclusionai/ling-2.6-1t warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3869 ms | 3869 ms | 3869 ms | 3869 ms | — | — | 1 |
| — | deepseek/deepseek-v4-flash-0731 warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3879 ms | 3879 ms | 3878 ms | 3878 ms | 84 tok/s warming | — | 3 |
| — | moonshotai/kimi-k3 warming | fireworks | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3884 ms | 4822 ms | 3884 ms | 4822 ms | 54 tok/s warming | — | 5 |
| — | openai/gpt-oss-20b warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3936 ms | 3936 ms | 3936 ms | 3936 ms | — | — | 1 |
| — | x-ai/grok-4.3 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3953 ms | 3953 ms | 3953 ms | 3953 ms | — | — | 1 |
| — | minimax/minimax-m3 warming | siliconflow | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3969 ms | 4149 ms | 3969 ms | 4149 ms | 51 tok/s warming | — | 4 |
| — | deepseek/deepseek-v3.2 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 3980 ms | 3980 ms | 3980 ms | 3980 ms | — | — | 1 |
| — | qwen/qwen3.5-397b-a17b warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4010 ms | 4080 ms | 4010 ms | 4080 ms | — | — | 2 |
| — | minimax/minimax-m2.1 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4025 ms | 4025 ms | 4024 ms | 4024 ms | — | — | 1 |
| — | openai/gpt-oss-120b warming | crusoe | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4033 ms | 4457 ms | 4033 ms | 4457 ms | 65 tok/s warming | — | 3 |
| — | qwen/qwen3.7-max warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4036 ms | 4513 ms | 4036 ms | 4513 ms | — | — | 2 |
| — | z-ai/glm-4.7-flash warming | cloudflare-workers-ai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4089 ms | 18454 ms | 4089 ms | 18454 ms | — | — | 3 |
| — | z-ai/glm-5 warming | siliconflow | 100.00% | 100.00% | 100.00% 45s | 100.00% | 4135 ms | 5176 ms | 4135 ms | 5175 ms | — | — | 6 |
| — | openai/gpt-oss-20b warming | parasail | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4137 ms | 5333 ms | 4137 ms | 5332 ms | — | 1 probe_config_error |
4 |
| — | qwen/qwen3.7-max warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4141 ms | 4141 ms | 4141 ms | 4141 ms | — | — | 1 |
| — | deepseek/deepseek-v3.2 warming | chutes | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4166 ms | 6378 ms | 4166 ms | 6378 ms | — | — | 7 |
| — | openai/gpt-4-turbo warming | openai | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4357 ms | 4357 ms | 4357 ms | 4357 ms | — | — | 1 |
| — | google/gemma-4-31b-it warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4371 ms | 4371 ms | 4371 ms | 4371 ms | — | — | 1 |
| — | deepseek/deepseek-v3-0324 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4453 ms | 4453 ms | 4453 ms | 4453 ms | — | — | 1 |
| — | moonshotai/kimi-k3 warming | gmi | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4554 ms | 7035 ms | 4554 ms | 7035 ms | 21 tok/s warming | — | 8 |
| — | google/gemini-3.1-pro-preview warming | google-ai-studio | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4624 ms | 5496 ms | 4624 ms | 5495 ms | 79 tok/s warming | — | 7 |
| — | deepseek/deepseek-v4-pro warming | digitalocean | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4645 ms | 5364 ms | 4645 ms | 5363 ms | 9 tok/s warming | — | 5 |
| — | z-ai/glm-5v-turbo warming | venice | 100.00% | 100.00% | 100.00% 45s | 100.00% | 4670 ms | 30080 ms | 4670 ms | 30079 ms | — | — | 6 |
| — | google/gemini-2.5-pro warming | google-vertex | 100.00% | 100.00% | 100.00% 20s | 100.00% | 4722 ms | 7429 ms | 4722 ms | 7429 ms | — | — | 9 |
| — | moonshotai/kimi-k3 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 5104 ms | 5104 ms | 5104 ms | 5104 ms | 25 tok/s warming | — | 1 |
| — | anthropic/claude-fable-5 warming | anthropic | 100.00% | 100.00% | 100.00% 20s | 100.00% | 5624 ms | 5624 ms | 5624 ms | 5624 ms | — | — | 1 |
| — | xiaomi/mimo-v2.5 warming | atlas-cloud | 100.00% | 100.00% | 100.00% 20s | 100.00% | 6380 ms | 6380 ms | 6380 ms | 6380 ms | — | — | 1 |
| — | z-ai/glm-4.7-flash warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 7660 ms | 7660 ms | 7659 ms | 7659 ms | — | — | 1 |
| — | minimax/minimax-m2.7 warming | deepinfra | 100.00% | 100.00% | 100.00% 20s | 100.00% | 8778 ms | 11330 ms | 8778 ms | 11329 ms | — | — | 3 |
| — | qwen/qwen3.7-plus-2026-05-26 warming | alibaba | 100.00% | 100.00% | 100.00% 20s | 100.00% | 11343 ms | 11343 ms | 11343 ms | 11343 ms | — | — | 1 |
| — | moonshotai/kimi-k3 warming | novita | 100.00% | 100.00% | 100.00% 20s | 100.00% | 19657 ms | 19657 ms | 19657 ms | 19657 ms | 25 tok/s warming | — | 1 |
| — | openai/gpt-oss-120b warming | siliconflow | 100.00% | 100.00% | 0.00% 20s | 100.00% | 26205 ms | 26205 ms | 26204 ms | 26204 ms | — | — | 1 |
| — | openai/gpt-oss-120b warming | tinfoil | 88.89% | 88.89% | 88.89% 20s | 100.00% | 1170 ms | 3674 ms | 1170 ms | 3673 ms | 84 tok/s warming | — | 9 |
| — | moonshotai/kimi-k2.6 warming | tinfoil | 88.89% | 88.89% | 88.89% 20s | 100.00% | 2213 ms | 4640 ms | 2213 ms | 4640 ms | — | — | 9 |
| — | z-ai/glm-4.7 warming | zai | 88.89% | 88.89% | 77.78% 20s | 100.00% | 2567 ms | 28741 ms | 2567 ms | 28740 ms | — | — | 9 |
| — | qwen/qwen3.5-397b-a17b warming | chutes | 80.00% | 80.00% | 80.00% 20s | 80.00% | 2460 ms | 3800 ms | 2460 ms | 3800 ms | — | — | 5 |
| — | nvidia/nemotron-3-120b-a12b warming | cloudflare-workers-ai | 75.00% | 75.00% | 75.00% 20s | 100.00% | 2787 ms | 13273 ms | 2786 ms | 13272 ms | — | — | 4 |
| — | zai-org/GLM-5.1 warming | nebius | 75.00% | 75.00% | 75.00% 45s | 100.00% | 6593 ms | 7934 ms | 6593 ms | 7933 ms | — | — | 4 |
| — | google/gemma-4-31b-turbo warming | chutes | 75.00% | 75.00% | 75.00% 20s | 100.00% | 10263 ms | 15854 ms | 10263 ms | 15854 ms | — | — | 4 |
| — | z-ai/glm-5 warming | chutes | 50.00% | 50.00% | 50.00% 45s | 100.00% | 1592 ms | 1592 ms | 1592 ms | 1592 ms | — | 1 unsupported_route |
2 |
| — | deepseek/deepseek-v4-pro warming | parasail | 50.00% | 50.00% | 50.00% 20s | 50.00% | 3537 ms | 3537 ms | 3537 ms | 3537 ms | — | — | 2 |
| — | moonshotai/kimi-k3 warming | siliconflow | 50.00% | 50.00% | 50.00% 20s | 50.00% | 5081 ms | 5081 ms | 5081 ms | 5081 ms | 17 tok/s warming | — | 2 |
| — | z-ai/glm-5.2 warming | digitalocean | 50.00% | 50.00% | 50.00% 45s | 50.00% | 5351 ms | 5351 ms | 5351 ms | 5351 ms | 28 tok/s warming | — | 2 |
| — | minimax/minimax-m3 warming | parasail | 50.00% | 50.00% | 0.00% 20s | 50.00% | 26905 ms | 26905 ms | 26905 ms | 26905 ms | 55 tok/s warming | — | 2 |
| — | moonshotai/kimi-k2.7-code warming | baseten | 40.00% | 40.00% | 40.00% 20s | 100.00% | 567 ms | 1586 ms | 567 ms | 1586 ms | 74 tok/s warming | — | 5 |
| — | qwen/qwen3-vl-8b-instruct warming | parasail | 33.33% | 33.33% | 33.33% 20s | 33.33% | 3880 ms | 3880 ms | 3880 ms | 3880 ms | — | — | 3 |
| — | qwen/qwen3-235b-a22b-2507 warming | crusoe | 25.00% | 25.00% | 25.00% 20s | 25.00% | 801 ms | 801 ms | 801 ms | 801 ms | — | — | 4 |
| — | baidu/cobuddy warming | novita | 0.00% | 0.00% | 0.00% 20s | 0.00% | — | — | — | — | — | — | 1 |
| — | deepseek/deepseek-v4-flash warming | atlas-cloud | — | — | — | — | — | — | — | — | 75 tok/s warming | — | 0 |
| — | deepseek/deepseek-v4-flash-0731 warming | atlas-cloud | 0.00% | 0.00% | 0.00% 20s | 100.00% | — | — | — | — | — | — | 2 |
| — | deepseek/deepseek-v4-flash-0731 warming | deepinfra | — | — | — | — | — | — | — | — | 63 tok/s warming | — | 0 |
| — | deepseek/deepseek-v4-flash-0731 warming | novita | — | — | — | — | — | — | — | — | 34 tok/s warming | — | 0 |
| — | deepseek/deepseek-v4-pro warming | alibaba | — | — | — | — | — | — | — | — | 40 tok/s warming | — | 0 |
| — | deepseek/deepseek-v4-pro warming | deepinfra | — | — | — | — | — | — | — | — | 73 tok/s warming | — | 0 |
| — | deepseek/deepseek-v4-pro warming | novita | — | — | — | — | — | — | — | — | 50 tok/s warming | — | 0 |
| — | deepseek/deepseek-v4-pro warming | together | — | — | — | — | — | — | — | — | 69 tok/s warming | 6 probe_config_error |
0 |
| — | google/gemini-3.1-pro-preview warming | atlas-cloud | — | — | — | — | — | — | — | — | 66 tok/s warming | — | 0 |
| — | google/gemma-3-12b-it warming | novita | 0.00% | 0.00% | 0.00% 20s | 100.00% | — | — | — | — | — | — | 2 |
| — | google/gemma-4-31b-it warming | deepinfra | — | — | — | — | — | — | — | — | 67 tok/s warming | — | 0 |
| — | google/gemma-4-31b-it warming | novita | — | — | — | — | — | — | — | — | 20 tok/s warming | — | 0 |
| — | google/gemma-4-31b-it warming | together | — | — | — | — | — | — | — | — | 64 tok/s warming | 1 probe_config_error |
0 |
| — | inclusionai/ling-2.6-flash warming | novita | 0.00% | 0.00% | 0.00% 20s | 0.00% | — | — | — | — | — | — | 1 |
| — | inclusionai/ling-3.0-flash warming | novita | — | — | — | — | — | — | — | — | 68 tok/s warming | — | 0 |
| — | meta-llama/llama-3.3-70b-instruct warming | novita | — | — | — | — | — | — | — | — | 37 tok/s warming | — | 0 |
| — | mindai/macaron-v1-tall warming | novita | 0.00% | 0.00% | 0.00% 20s | 100.00% | — | — | — | — | — | — | 1 |
| — | mindai/macaron-v1-venti warming | novita | — | — | — | — | — | — | — | — | 12 tok/s warming | — | 0 |
| — | minimax/minimax-m3 warming | deepinfra | — | — | — | — | — | — | — | — | 19 tok/s warming | — | 0 |
| — | minimax/minimax-m3 warming | together | — | — | — | — | — | — | — | — | 61 tok/s warming | 7 probe_config_error |
0 |
| — | moonshotai/kimi-k2.6 warming | deepinfra | — | — | — | — | — | — | — | — | 37 tok/s warming | — | 0 |
| — | nvidia/nemotron-3-ultra-550b-a55b warming | together | — | — | — | — | — | — | — | — | 124 tok/s warming | — | 0 |
| — | openai/gpt-4.1-mini warming | atlas-cloud | — | — | — | — | — | — | — | — | 83 tok/s warming | — | 0 |
| — | openai/gpt-5.6-luna warming | atlas-cloud | — | — | — | — | — | — | — | — | 64 tok/s warming | — | 0 |
| — | qwen/qwen-plus-character warming | alibaba | — | — | — | — | — | — | — | — | 33 tok/s warming | — | 0 |
| — | qwen/qwen3-vl-flash-2026-01-22 warming | alibaba | 0.00% | 0.00% | 0.00% 20s | 100.00% | — | — | — | — | — | — | 3 |
| — | qwen/qwen3.5-122b-a10b warming | deepinfra | 0.00% | 0.00% | 0.00% 20s | 0.00% | — | — | — | — | — | — | 2 |
| — | qwen/qwen3.7-flash warming | alibaba | — | — | — | — | — | — | — | — | 114 tok/s warming | — | 0 |
| — | tencent/hy3 warming | atlas-cloud | — | — | — | — | — | — | — | — | 111 tok/s warming | — | 0 |
| — | thedrummer/skyfall-36b-v2 warming | parasail | 0.00% | 0.00% | 0.00% 20s | 0.00% | — | — | — | — | — | — | 3 |
| — | thinkingmachines/inkling warming | deepinfra | — | — | — | — | — | — | — | — | 67 tok/s warming | — | 0 |
| — | x-ai/grok-4.5 warming | atlas-cloud | — | — | — | — | — | — | — | — | 55 tok/s warming | — | 0 |
| — | xiaomi/mimo-v2.5-pro warming | digitalocean | — | — | — | — | — | — | — | — | 64 tok/s warming | — | 0 |
| — | z-ai/glm-5.2 warming | alibaba | — | — | — | — | — | — | — | — | 37 tok/s warming | — | 0 |
| — | z-ai/glm-5.2 warming | deepinfra | — | — | — | — | — | — | — | — | 49 tok/s warming | — | 0 |
| — | z-ai/glm-5.2 warming | novita | — | — | — | — | — | — | — | — | 25 tok/s warming | — | 0 |