Fireworks AI
Explore Fireworks AI models on TrustedRouter with current routes, token pricing, policy sources, privacy posture, regional availability, and API support.
Fireworks AIfireworks
No provider claim
| Provider | Fireworks AI |
|---|---|
| Provider website | https://fireworks.ai/ |
| Models | 12 public models |
| Prepaid routes | 12 |
| BYOK routes | 11 |
| Zero data retention | not claimed |
| Confidential compute | not claimed |
| Provider E2EE | not claimed |
| Policy note | No provider-ZDR claim is tracked here. Fireworks publishes security, privacy, and zero-retention documentation; enable a contracted ZDR posture before marking this provider as ZDR. Policy source |
Measured performance
40 samplesContinuously sampled across Fireworks 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 | 2684 ms |
|---|---|
| Effective throughput | 50 tok/s n=13 |
| Uptime | 100.00% |
| Model | p50 TTFT | p50 TTFB | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|---|
| openai/gpt-oss-20b | 733 ms | 733 ms | — | 100.00% | — | 2 |
| openai/gpt-oss-120b | 827 ms | 827 ms | 150 tok/s n=1 | 100.00% | — | 2 |
| deepseek/deepseek-v4-flash-0731 | 1362 ms | 1362 ms | 46 tok/s n=1 | 100.00% | — | 2 |
| moonshotai/kimi-k2.6 | 1472 ms | 1472 ms | 44 tok/s n=1 | 100.00% | — | 3 |
| deepseek/deepseek-v4-pro-0423 | 1970 ms | 1970 ms | 53 tok/s n=2 | 100.00% | — | 3 |
| deepseek/deepseek-v4-pro | 2205 ms | 2204 ms | — | 100.00% | — | 5 |
| moonshotai/kimi-k2.7-code | 2566 ms | 2566 ms | 73 tok/s n=1 | 100.00% | — | 2 |
| minimax/minimax-m3 | 2684 ms | 2684 ms | 82 tok/s n=2 | 100.00% | — | 7 |
| z-ai/glm-5.2-fast | 2801 ms | 2800 ms | 48 tok/s n=1 | 100.00% | — | 5 |
| z-ai/glm-5.2 | 3676 ms | 3675 ms | 50 tok/s n=2 | 100.00% | — | 3 |
| minimax/minimax-m2.7 | 3907 ms | 3907 ms | — | 100.00% | — | 2 |
| moonshotai/kimi-k3 | 7290 ms | 7290 ms | 49 tok/s n=2 | 100.00% | — | 4 |
Fireworks AI performance history · Full provider & model leaderboard.
Models served by Fireworks AI.
Each row links to pricing, provider, benchmark, and API pages for the model.
| Model | AI IQ | Context | Endpoints | Prompt | Completion | Routes |
|---|---|---|---|---|---|---|
deepseek/deepseek-v4-flash-0731DeepSeek: DeepSeek V4 Flash 0731 |
— | 1,048,576 | 2 | $0.1477/1M | $0.2954/1M | prepaid BYOK |
deepseek/deepseek-v4-proDeepSeek: DeepSeek V4 Pro |
IQ 117#31 | 1,048,576 | 2 | $1.8357/1M | $3.6714/1M | prepaid BYOK |
deepseek/deepseek-v4-pro-0423DeepSeek V4 Pro 0423 |
— | 1,048,576 | 1 | $1.8357/1M | $3.6714/1M | prepaid |
minimax/minimax-m2.7MiniMax: MiniMax M2.7 |
IQ 109#50 | 204,800 | 2 | $0.3165/1M | $1.266/1M | prepaid BYOK |
minimax/minimax-m3MiniMax: MiniMax M3 |
IQ 114#39 | 1,048,576 | 2 | $0.3165/1M | $1.266/1M | prepaid BYOK |
moonshotai/kimi-k2.6MoonshotAI: Kimi K2.6 |
IQ 119#23 | 262,144 | 2 | $1.00225/1M | $4.22/1M | prepaid BYOK |
moonshotai/kimi-k2.7-codeMoonshotAI: Kimi K2.7 Code |
IQ 118#28 | 262,144 | 2 | $1.00225/1M | $4.22/1M | prepaid BYOK |
moonshotai/kimi-k3MoonshotAI: Kimi K3 |
IQ 123#16 | 1,048,576 | 2 | $3.165/1M | $15.825/1M | prepaid BYOK |
openai/gpt-oss-120bOpenAI: gpt-oss-120b |
IQ 105#61 | 131,072 | 2 | $0.15825/1M | $0.633/1M | prepaid BYOK |
openai/gpt-oss-20bOpenAI: gpt-oss-20b |
IQ 100#77 | 131,072 | 2 | $0.07385/1M | $0.3165/1M | prepaid BYOK |
z-ai/glm-5.2Z.ai: GLM 5.2 |
IQ 120#19 | 1,048,576 | 2 | $1.477/1M | $4.642/1M | prepaid BYOK |
z-ai/glm-5.2-fastGLM 5.2 Fast on Fireworks |
— | 1,048,576 | 2 | $2.954/1M | $9.284/1M | prepaid BYOK |
Questions
Does Fireworks AI have zero data retention?
TrustedRouter does not currently mark Fireworks 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 Fireworks AI end-to-end encrypted?
TrustedRouter does not currently mark Fireworks 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 Fireworks AI models are available through TrustedRouter?
This page currently lists 12 public Fireworks AI models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.