Pearl Research Labs
Try Pearl Research Labs on us. Starter credit covers your first requests, then one key reaches all 6 Pearl Research Labs models, with the privacy posture shown on every route.
Pearl Research Labspearl
These privacy labels describe Pearl Research Labs, the upstream model provider. Confidential requires all three: verified provider compute, end-to-end encryption and explicit zero data retention. Account and billing metadata are outside the content-retention claim.
| Provider | Pearl Research Labs |
|---|---|
| API operator country | IL Legal home of the API operator, not an inference-location guarantee. |
| Headquarters | Tel Aviv-Yafo, Israel Provider-submitted operating and registered address; public policy for contact. Source; reviewed 2026-09-27. |
| Routing status | Active |
| Provider website | https://pearlresearch.ai/ |
| Models | 6 public models |
| Credits routes | 6 |
| Zero data retention | no |
| Verified confidential inference | Not verified |
| Policy note | Pearl Research states that it retains operational metadata for security and does not train on customer data. No public contractual zero-data-retention terms are linked, so TrustedRouter classifies these routes as Standard and excludes them from ZDR, confidential-compute, and E2EE routing. Policy source |
Inference locations
Provider API declarations are also available on each model's endpoint API. Direct chat and Responses replies separate advertised availability from serving location in usage.inference_location; the actual serving region remains unknown when not reported.
usage.provider_usage.gateway_region identifies the TrustedRouter gateway, not the provider's inference location. A value such as europe-west4 does not mean the model ran in Europe. usage.inference_location.serving_region remains null when the provider has not reported it. Older gateways may return the gateway-only field as provider_usage.region.
Where the upstream GPUs execute the model, not the provider's headquarters, API ingress, or data storage. ZDR and confidential inference do not by themselves restrict geography.
- Declared countries or regions
- United States (Central); Taiwan; Australia
Pearl supplied these serving regions in its marketplace application. This is provider-wide information, not a confirmed per-model or exhaustive failover list. GLM 5.3 Flash and DeepSeek V4.1 Flash do not declare regions in the reviewed native catalog. - Can locations change?
- Per-model placement and cross-region failover behavior require provider confirmation.
- Provider region pinning
- No per-request region-pinning contract verified in the reviewed API information.
- Pinning through TrustedRouter
- Not supported by the current TrustedRouter integration.
- Infrastructure declaration
- Multi-region provider: US Central, Taiwan and Australia declared; model-specific inference location and exhaustive failover scope unconfirmed.
- Evidence
- Provider-submitted marketplace information; no public per-model residency declaration. Reviewed .
Provider selection alone is not a region pin. TrustedRouter's US provider filter selects company jurisdiction, not GPU location. A regional gateway hostname does not pin a downstream GPU. For a binding residency requirement, confirm the model, region, failover scope and enforced controls before sending sensitive data.
Measured performance
49 samplesContinuously sampled across Pearl Research Labs'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 | 1520 ms |
|---|---|
| Effective throughput | 186 tok/s n=3 |
| Uptime | 100.00% |
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| deepseek/deepseek-v4.1-flash | 1520 ms | 87 tok/s n=1 | 100.00% | — | 33 |
| z-ai/glm-5.3-flash | 1195 ms | — | 100.00% | — | 2 |
| deepseek/deepseek-v4.1-flash-1017 | 1554 ms | — | 100.00% | — | 4 |
| z-ai/glm-5.3 | 1645 ms | 186 tok/s n=2 | 100.00% | — | 2 |
| qwen/qwen3.8-27b | 2230 ms | — | 100.00% | — | 5 |
| qwen/qwen3.8-flash-next | 2691 ms | — | 100.00% | — | 3 |
Pearl Research Labs performance history · Full provider & model leaderboard.
Models served by Pearl Research Labs.
Each row links to pricing, provider, benchmark, and API pages for the model.
| Model | AI IQ | Context | Input | Cached input | Output |
|---|---|---|---|---|---|
deepseek/deepseek-v4.1-flashDeepSeek: DeepSeek V4.1 Flash |
IQ 126#24 | 1,048,576 | $0.05275/1M | $0.01/1M | $0.3165/1M |
deepseek/deepseek-v4.1-flash-1017DeepSeek: DeepSeek V4.1 Flash 1017 |
IQ 128#19 | 1,048,576 | $0.05275/1M | $0.01/1M | $0.3165/1M |
qwen/qwen3.8-27bQwen: Qwen3.8 27B |
IQ 115#57 | 65,536 | $0.211/1M | $0.05275/1M | $2.11/1M |
qwen/qwen3.8-flash-nextQwen: Qwen3.8-Flash-Next |
— | 262,144 | $0.1477/1M | $0.015825/1M | $0.47475/1M |
z-ai/glm-5.3Z.ai: GLM 5.3 |
IQ 124#30 | 1,048,576 | $1.1605/1M | $0.24265/1M | $3.6925/1M |
z-ai/glm-5.3-flashZ.ai: GLM 5.3 Flash |
IQ 119#50 | 1,048,576 | $0.07385/1M | $0.015825/1M | $0.2532/1M |
Questions
Does Pearl Research Labs have zero data retention?
TrustedRouter does not currently mark Pearl Research Labs 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 Pearl Research Labs end-to-end encrypted?
TrustedRouter does not currently mark Pearl Research Labs 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 Pearl Research Labs models are available through TrustedRouter?
This page currently lists 6 public Pearl Research Labs models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.