OpenAI compatible API · Attested · Public status

FriendliAI

Explore FriendliAI 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.

FriendliAIfriendli

No provider claim

All providers

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

ProviderFriendliAI
API operator countryUS
Legal home of the API operator, not an inference-location guarantee.
HeadquartersNot verified
Routing statusActive
Provider websitehttps://friendli.ai/
Models7 public models
Credits routes7
Zero data retentionnot claimed
Verified confidential inferenceNot verified
Policy noteNo provider-ZDR claim is tracked here. Friendli's legal/privacy terms are linked for users who need to review API data handling.
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.

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
Not verified
No inference-location declaration has been verified for this provider.
Can locations change?
Not verified. Do not assume that the serving location is fixed.
Provider region pinning
Not verified.
Pinning through TrustedRouter
No verified provider-specific location pin documented here.
Infrastructure declaration
Inference location not verified; no country-specific residency commitment.
Evidence
Not yet 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

27 samples

Continuously sampled across FriendliAI'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 TTFT1403 ms
Effective throughput70 tok/s n=4
Uptime100.00%
Modelp50 TTFTEffective throughputUptimeConfig excludedAvailability samples
z-ai/glm-5.1 783 ms — 100.00% — 5
google/gemma-4-31b-it 962 ms 65 tok/s n=2 100.00% — 5
minimax/minimax-m2.5 1102 ms — 100.00% — 1
z-ai/glm-5.3 1403 ms 214 tok/s n=1 100.00% — 3
deepseek/deepseek-v3.2 1499 ms — 100.00% — 4
z-ai/glm-5.3-flash 9853 ms — 100.00% 1 router_error 9
z-ai/glm-5.2 — 75 tok/s n=1 — — 0

FriendliAI performance history · Full provider & model leaderboard.

Provider models

Models served by FriendliAI.

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

Model AI IQ Context Input Cached input Output
deepseek/deepseek-v3.2
DeepSeek: DeepSeek V3.2
IQ 103#89 131,072 $0.5275/1M $0.26375/1M $1.5825/1M
google/gemma-4-31b-it
Google: Gemma 4 31B
IQ 101#99 262,144 $0.1477/1M Not published $0.422/1M
minimax/minimax-m2.5
MiniMax: MiniMax M2.5
IQ 106#80 196,608 $0.3165/1M $0.0633/1M $1.266/1M
z-ai/glm-5.1
Z.ai: GLM 5.1
IQ 113#56 202,752 $1.477/1M $0.2743/1M $4.642/1M
z-ai/glm-5.2
Z.ai: GLM 5.2
IQ 119#40 1,048,576 $1.477/1M $0.2743/1M $4.642/1M
z-ai/glm-5.3
Z.ai: GLM 5.3
IQ 127#16 1,048,576 $1.3293/1M $0.24687/1M $4.1778/1M
z-ai/glm-5.3-flash
Z.ai: GLM 5.3 Flash
IQ 120#39 1,048,576 $0.15825/1M $0.03165/1M $0.5275/1M

Questions

Does FriendliAI have zero data retention?

TrustedRouter does not currently mark FriendliAI 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 FriendliAI end-to-end encrypted?

TrustedRouter does not currently mark FriendliAI 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 FriendliAI models are available through TrustedRouter?

This page currently lists 7 public FriendliAI models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.

Workspace access

Sign in

Choose a sign in method to access your TrustedRouter workspace.

By signing in you agree to the terms of service and privacy policy.