OpenAI compatible API · Attested · Public status

Novita AI

Explore Novita AI models on TrustedRouter with current routes, token pricing, policy sources, privacy posture, regional availability, and API support.

Verify gateway

Onebase URL to migrate

100sof models and routes

0prompt or output logs. Always.

Novita AInovita

No provider claim

All providers

These privacy labels describe Novita AI, the upstream model provider. ZDR is a retention policy; verified confidential inference additionally requires attested provider compute and end-to-end encryption.

Provider	Novita AI
Routing status	Active
Provider website	https://novita.ai/
Models	93 public models
Credits routes	93
Zero data retention	not claimed
Verified confidential inference	Not verified
Policy note	No provider-ZDR claim is tracked here. Novita's privacy policy says personal information is not used for model training; customer-content processing is governed by customer agreements. Policy source

Measured performance

30 samples

Continuously sampled across Novita AI's routed models: p50 TTFT, effective throughput, and success rate. Effective throughput uses provider-reported output tokens over complete request time. Unsupported route and probe-configuration rows are separated from provider downtime. No prompt or output content stored.

p50 TTFT	2004 ms
Effective throughput	78 tok/s n=9
Uptime	83.33%

Model	p50 TTFT	Effective throughput	Uptime	Config excluded	Availability samples
deepseek/deepseek-ocr-2	703 ms	—	100.00%	—	4
sao10k/l31-70b-euryale-v2.2	1207 ms	—	100.00%	—	1
moonshotai/kimi-k2.6	1426 ms	29 tok/s n=1	100.00%	—	2
zai-org/autoglm-phone-9b-multilingual	1430 ms	—	100.00%	—	1
qwen/qwen3-next-80b-a3b-instruct	1573 ms	—	100.00%	—	1
zai-org/glm-4.6v	1584 ms	—	100.00%	—	1
qwen/qwen3.8-27b	1668 ms	—	100.00%	—	1
deepseek/deepseek-v4.1-flash	1863 ms	—	100.00%	—	1
meta-llama/llama-3.3-70b-instruct	2004 ms	—	100.00%	—	1
nvidia/nemotron-3-nano-30b-a3b	2022 ms	—	100.00%	—	1
qwen/qwen3-vl-30b-a3b-instruct	2117 ms	—	100.00%	—	1
minimax/minimax-m3	2189 ms	87 tok/s n=2	100.00%	—	1
qwen/qwen3.8-max	2466 ms	—	100.00%	—	2
z-ai/glm-5.2	2827 ms	—	100.00%	—	1
meta-llama/llama-4-scout-17b-16e-instruct	3040 ms	—	100.00%	—	1
minimax/minimax-m2.7	3817 ms	—	100.00%	—	1
zai-org/glm-4.7	4556 ms	—	100.00%	—	1
baidu/ernie-4.5-vl-424b-a47b	5075 ms	—	100.00%	—	1
deepseek/deepseek-v4-pro	5524 ms	—	100.00%	—	1
openai/gpt-oss-120b	22168 ms	6 tok/s n=1	100.00%	—	1
baichuan/baichuan-m2-32b	—	—	0.00%	—	2
baidu/ernie-4.5-21B-a3b	—	—	0.00%	—	1
deepseek/deepseek-v4-flash	—	95 tok/s n=1	—	—	0
google/gemma-4-31b-it	—	78 tok/s n=2	—	—	0
mindai/macaron-v1-tall	—	—	0.00%	—	1
moonshotai/kimi-k3	—	45 tok/s n=1	—	—	0
qwen/qwen3-235b-a22b-thinking-2507	—	101 tok/s n=1	—	—	0
zai-org/glm-4.7-flash	—	—	0.00%	—	1

Novita AI performance history · Full provider & model leaderboard.

Provider models

Models served by Novita AI.

Each row links to pricing, provider, benchmark, and API pages for the model.

Model	AI IQ	Context	Input	Cached input	Output
`Sao10K/L3-8B-Stheno-v3.2` Llama 3 8B Stheno V3.2	—	8,192	$0.05275/1M	Not published	$0.05275/1M
`baichuan/baichuan-m2-32b` Baichuan M2 32B	—	131,072	$0.07385/1M	Not published	$0.07385/1M
`baidu/cobuddy` CoBuddy	—	131,072	$0.2954/1M	$0.07385/1M	$1.19215/1M
`baidu/ernie-4.5-21B-a3b` ERNIE 4.5 21B A3B	—	120,000	$0.07385/1M	Not published	$0.2954/1M
`baidu/ernie-4.5-vl-424b-a47b` Baidu: ERNIE 4.5 VL 424B A47B	—	123,000	$0.4431/1M	Not published	$1.31875/1M
`deepseek/deepseek-ocr-2` DeepSeek OCR 2	—	8,192	$0.03165/1M	Not published	$0.03165/1M
`deepseek/deepseek-r1-0528` DeepSeek: R1 0528	—	163,840	$0.7385/1M	$0.36925/1M	$2.6375/1M
`deepseek/deepseek-r1-distill-llama-70b` DeepSeek: R1 Distill Llama 70B	—	8,192	$0.844/1M	Not published	$0.844/1M
`deepseek/deepseek-r1-turbo` DeepSeek R1 Turbo	—	64,000	$0.7385/1M	Not published	$2.6375/1M
`deepseek/deepseek-v3.1` DeepSeek V3.1	IQ 95#105	131,072	$0.28485/1M	$0.142425/1M	$1.055/1M
`deepseek/deepseek-v3.1-terminus` DeepSeek: DeepSeek V3.1 Terminus	—	131,072	$0.28485/1M	$0.142425/1M	$1.055/1M
`deepseek/deepseek-v3.2` DeepSeek: DeepSeek V3.2 benchmarks	IQ 103#77	163,840	$0.283795/1M	$0.141898/1M	$0.422/1M
`deepseek/deepseek-v3.2-exp` DeepSeek: DeepSeek V3.2 Exp	—	163,840	$0.28485/1M	Not published	$0.43255/1M
`deepseek/deepseek-v4-flash` DeepSeek: DeepSeek V4 Flash 0423 benchmarks	IQ 116#37	1,048,576	$0.1477/1M	$0.02954/1M	$0.2954/1M
`deepseek/deepseek-v4-flash-0731` DeepSeek: DeepSeek V4 Flash 0731	—	1,048,576	$0.4642/1M	$0.02954/1M	$1.3926/1M
`deepseek/deepseek-v4-flash-vision-exp` DeepSeek: DeepSeek V4 Flash Vision Exp	IQ 115#42	1,048,576	$0.4642/1M	$0.02954/1M	$1.3926/1M
`deepseek/deepseek-v4-pro` DeepSeek: DeepSeek V4 Pro 0423 benchmarks	IQ 114#47	1,048,576	$1.688/1M	$0.142425/1M	$3.376/1M
`deepseek/deepseek-v4.1-flash` DeepSeek: DeepSeek V4.1 Flash	IQ 116#38	1,048,576	$0.3165/1M	$0.01/1M	$1.266/1M
`google/gemma-3-12b-it` Google: Gemma 3 12B	—	131,072	$0.05275/1M	Not published	$0.1055/1M
`google/gemma-3-27b-it` Google: Gemma 3 27B	—	131,072	$0.125545/1M	Not published	$0.211/1M
`google/gemma-4-26b-a4b-it` Google: Gemma 4 26B A4B	IQ 100#89	262,144	$0.13715/1M	Not published	$0.422/1M
`google/gemma-4-31b-it` Google: Gemma 4 31B	IQ 103#80	262,144	$0.1477/1M	Not published	$0.422/1M
`inclusionai/ling-3.0-flash` inclusionAI: Ling 3.0 Flash	IQ 94#109	131,072	$0.0633/1M	$0.01266/1M	$0.1899/1M
`meta-llama/llama-3.1-8b-instruct` Meta: Llama 3.1 8B Instruct	—	131,072	$0.0211/1M	Not published	$0.05275/1M
`meta-llama/llama-3.3-70b-instruct` Meta: Llama 3.3 70B Instruct benchmarks	—	131,072	$0.142425/1M	Not published	$0.422/1M
`meta-llama/llama-4-maverick-17b-128e-instruct-fp8` Llama 4 Maverick Instruct	—	1,048,576	$0.28485/1M	Not published	$0.89675/1M
`meta-llama/llama-4-scout-17b-16e-instruct` Llama 4 Scout Instruct	—	131,072	$0.1899/1M	Not published	$0.62245/1M
`microsoft/wizardlm-2-8x22b` WizardLM-2 8x22B	—	65,535	$0.6541/1M	Not published	$0.6541/1M
`mindai/macaron-v1-tall` Macaron V1 Tall	—	262,144	$0.261113/1M	$0.04642/1M	$1.50865/1M
`mindai/macaron-v1-venti` Macaron V1 Venti	—	1,048,576	$0.870375/1M	$0.174075/1M	$2.611125/1M
`minimax/minimax-m2` MiniMax: MiniMax M2	—	204,800	$0.3165/1M	$0.03165/1M	$1.266/1M
`minimax/minimax-m2.1` MiniMax: MiniMax M2.1	IQ 101#87	204,800	$0.3165/1M	$0.03165/1M	$1.266/1M
`minimax/minimax-m2.5` MiniMax: MiniMax M2.5	IQ 106#74	196,608	$0.3165/1M	$0.03165/1M	$1.266/1M
`minimax/minimax-m2.5-highspeed` MiniMax M2.5 Highspeed	—	204,800	$0.633/1M	$0.03165/1M	$2.532/1M
`minimax/minimax-m2.7` MiniMax: MiniMax M2.7	IQ 107#72	196,608	$0.3165/1M	$0.0633/1M	$1.266/1M
`minimax/minimax-m3` MiniMax: MiniMax M3 benchmarks	IQ 115#45	524,288	$0.3165/1M	$0.0633/1M	$1.266/1M
`minimaxai/minimax-m1-80k` MiniMax M1	—	1,000,000	$0.58025/1M	Not published	$2.321/1M
`mistralai/mistral-nemo` Mistral: Mistral Nemo	—	131,072	$0.0422/1M	Not published	$0.17935/1M
`moonshotai/kimi-k2-0905` MoonshotAI: Kimi K2 0905	—	262,144	$0.633/1M	Not published	$2.6375/1M
`moonshotai/kimi-k2-instruct` Kimi K2 Instruct	IQ 91#116	131,072	$0.60135/1M	Not published	$2.4265/1M
`moonshotai/kimi-k2-thinking` MoonshotAI: Kimi K2 Thinking	IQ 91#116	262,144	$0.633/1M	$0.15825/1M	$2.6375/1M
`moonshotai/kimi-k2.5` MoonshotAI: Kimi K2.5	IQ 111#57	262,144	$0.633/1M	$0.1055/1M	$3.165/1M
`moonshotai/kimi-k2.6` MoonshotAI: Kimi K2.6 benchmarks	IQ 118#35	262,144	$0.844/1M	$0.1688/1M	$3.587/1M
`moonshotai/kimi-k2.7-code` MoonshotAI: Kimi K2.7 Code benchmarks	IQ 108#66	262,144	$1.00225/1M	$0.20045/1M	$4.22/1M
`moonshotai/kimi-k3` MoonshotAI: Kimi K3	IQ 121#27	1,048,576	$3.165/1M	$0.3165/1M	$15.825/1M
`nvidia/nemotron-3-nano-30b-a3b` NVIDIA: Nemotron 3 Nano 30B A3B	—	262,144	$0.05275/1M	Not published	$0.211/1M
`openai/gpt-oss-120b` OpenAI: gpt-oss-120b	IQ 98#99	131,072	$0.05275/1M	Not published	$0.26375/1M
`openai/gpt-oss-20b` OpenAI: gpt-oss-20b	IQ 93#113	131,072	$0.0422/1M	Not published	$0.15825/1M
`qwen/qwen-2.5-72b-instruct` Qwen2.5 72B Instruct	—	32,000	$0.4009/1M	Not published	$0.422/1M
`qwen/qwen-mt-plus` Qwen MT Plus	—	16,384	$0.26375/1M	Not published	$0.79125/1M
`qwen/qwen3-235b-a22b-fp8` Qwen3 235B A22B	—	40,960	$0.211/1M	Not published	$0.844/1M
`qwen/qwen3-235b-a22b-instruct-2507` Qwen3 235B A22B Instruct 2507	—	131,072	$0.09495/1M	Not published	$0.6119/1M
`qwen/qwen3-235b-a22b-thinking-2507` Qwen: Qwen3 235B A22B Thinking 2507	—	131,072	$0.3165/1M	Not published	$3.165/1M
`qwen/qwen3-coder-30b-a3b-instruct` Qwen: Qwen3 Coder 30B A3B Instruct	—	160,000	$0.07385/1M	Not published	$0.28485/1M
`qwen/qwen3-coder-480b-a35b-instruct` Qwen3 Coder 480B A35B Instruct	—	262,144	$0.4009/1M	Not published	$1.63525/1M
`qwen/qwen3-coder-next` Qwen: Qwen3 Coder Next	—	262,144	$0.211/1M	Not published	$1.5825/1M
`qwen/qwen3-max` Qwen3 Max	—	262,144	$2.22605/1M	Not published	$8.91475/1M
`qwen/qwen3-next-80b-a3b-instruct` Qwen: Qwen3 Next 80B A3B Instruct	—	262,144	$0.15825/1M	Not published	$1.5825/1M
`qwen/qwen3-omni-30b-a3b-instruct` Qwen3 Omni 30B A3B Instruct	—	65,536	$0.26375/1M	Not published	$1.02335/1M
`qwen/qwen3-omni-30b-a3b-thinking` Qwen3 Omni 30B A3B Thinking	—	65,536	$0.26375/1M	Not published	$1.02335/1M
`qwen/qwen3-vl-235b-a22b-instruct` Qwen: Qwen3 VL 235B A22B Instruct	—	262,144	$0.3165/1M	Not published	$1.5825/1M
`qwen/qwen3-vl-235b-a22b-thinking` Qwen: Qwen3 VL 235B A22B Thinking	—	131,072	$1.0339/1M	Not published	$4.16725/1M
`qwen/qwen3-vl-30b-a3b-instruct` Qwen: Qwen3 VL 30B A3B Instruct	—	262,144	$0.211/1M	Not published	$0.7385/1M
`qwen/qwen3.5-122b-a10b` Qwen: Qwen3.5-122B-A10B	—	262,144	$0.422/1M	Not published	$3.376/1M
`qwen/qwen3.5-27b` Qwen: Qwen3.5-27B	—	262,144	$0.3165/1M	Not published	$2.532/1M
`qwen/qwen3.5-35b-a3b` Qwen: Qwen3.5-35B-A3B	—	262,144	$0.26375/1M	Not published	$2.11/1M
`qwen/qwen3.5-397b-a17b` Qwen: Qwen3.5 397B A17B	—	262,144	$0.633/1M	Not published	$3.798/1M
`qwen/qwen3.6-27b` Qwen: Qwen3.6 27B	IQ 108#68	262,144	$0.633/1M	Not published	$3.798/1M
`qwen/qwen3.6-35b-a3b` Qwen: Qwen3.6 35B A3B	IQ 100#94	262,144	$0.26164/1M	Not published	$1.566675/1M
`qwen/qwen3.7-max` Qwen3.7 Max	IQ 119#31	1,000,000	$1.31875/1M	$0.26375/1M	$3.95625/1M
`qwen/qwen3.8-2.4t-a95b` Qwen: Qwen3.8 2.4T A95B	IQ 122#24	262,144	$2.11/1M	$0.26375/1M	$6.33/1M
`qwen/qwen3.8-27b` Qwen: Qwen3.8 27B	IQ 110#60	262,144	$0.4431/1M	$0.089675/1M	$3.165/1M
`qwen/qwen3.8-flash` Qwen3.8 Flash	IQ 115#46	1,000,000	$0.15825/1M	$0.01688/1M	$0.49585/1M
`qwen/qwen3.8-max` Qwen3.8 Max	IQ 122#25	1,000,000	$2.11/1M	$0.26375/1M	$6.33/1M
`sao10k/l3-8b-lunaris` Llama 3 8B Lunaris	—	8,192	$0.05275/1M	Not published	$0.05275/1M
`sao10k/l31-70b-euryale-v2.2` Llama 3.1 70B Euryale V2.2	—	8,192	$1.5614/1M	Not published	$1.5614/1M
`stepfun/step-3.7-flash` StepFun: Step 3.7 Flash	IQ 102#84	262,144	$0.211/1M	$0.0422/1M	$1.21325/1M
`tencent/hy3` Tencent: Hy3	IQ 100#90	262,144	$0.1477/1M	$0.036925/1M	$0.6119/1M
`xiaomimimo/mimo-v2.5` Xiaomi MiMo V2.5	IQ 108#67	1,048,576	$0.17724/1M	$0.01/1M	$0.35448/1M
`xiaomimimo/mimo-v2.5-pro` Xiaomi MiMo V2.5 Pro	IQ 112#55	1,048,576	$0.55071/1M	$0.01/1M	$1.10142/1M
`z-ai/glm-5.2` Z.ai: GLM 5.2 benchmarks	IQ 120#28	1,048,576	$1.477/1M	$0.2743/1M	$4.642/1M
`z-ai/glm-5.3` Z.ai: GLM 5.3	IQ 123#18	1,048,576	$1.477/1M	$0.2743/1M	$4.642/1M
`z-ai/glm-5.3-flash` Z.ai: GLM 5.3 Flash	IQ 116#40	1,048,576	$0.15825/1M	$0.03165/1M	$0.5275/1M
`z-ai/glm-5.3-p` zai-org/glm-5.3-p	—	1,048,576	$1.477/1M	$0.2743/1M	$4.642/1M
`zai-org/autoglm-phone-9b-multilingual` AutoGLM Phone 9B Multilingual	—	65,536	$0.036925/1M	Not published	$0.14559/1M
`zai-org/glm-4.5-air` GLM 4.5 Air	—	131,072	$0.13715/1M	$0.026375/1M	$0.89675/1M
`zai-org/glm-4.5v` GLM 4.5V	—	65,536	$0.633/1M	$0.11605/1M	$1.899/1M
`zai-org/glm-4.6` GLM 4.6	—	204,800	$0.58025/1M	$0.11605/1M	$2.321/1M
`zai-org/glm-4.6v` GLM 4.6V	—	131,072	$0.3165/1M	$0.058025/1M	$0.9495/1M
`zai-org/glm-4.7` GLM 4.7	IQ 103#81	204,800	$0.633/1M	$0.11605/1M	$2.321/1M
`zai-org/glm-4.7-flash` GLM 4.7 Flash	—	200,000	$0.07385/1M	$0.01055/1M	$0.422/1M
`zai-org/glm-5` GLM 5	IQ 105#75	202,800	$1.055/1M	$0.211/1M	$3.376/1M
`zai-org/glm-5.1` GLM 5.1	IQ 113#49	204,800	$1.4559/1M	$0.2743/1M	$4.642/1M

Questions

Does Novita AI have zero data retention?

TrustedRouter does not currently mark Novita AI as provider-level zero data retention. Use trustedrouter/zdr or provider.min_privacy=zdr to select a different eligible route, and review the linked policy source for changes.

Is Novita AI end-to-end encrypted?

TrustedRouter does not currently mark Novita AI as end-to-end encrypted at the provider boundary. The TrustedRouter gateway is still attested, but the selected provider normally receives the request in order to run the model. Use trustedrouter/e2e for the stronger route requirement.

Which Novita AI models are available through TrustedRouter?

This page currently lists 93 public Novita AI models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.