Fireworks — pricing & performance
Models served
24
of 1,128 tracked models
Mean output speed
68
tokens per second, averaged over
12 measured endpoints
12 measured endpoints
Mean latency
1.3s
time to first token, averaged over
12 measured endpoints
12 measured endpoints
Cheapest blended
$0.12
per 1M tokens at a 3:1
input:output blend
input:output blend
Output speed on Fireworks
Output tokens per second for the 12 of 24 models on Fireworks with published throughput telemetry, the most recent figure OpenRouter publishes for each host. The full table below lists all 24, measured or not. Higher is better. We do not run these measurements ourselves.
The Known Good
24 models on Fireworks
Full leaderboard →
| Model | Creator | Input $/M | Output $/M | Blended | Speed | TTFT | Context |
|---|---|---|---|---|---|---|---|
DeepSeek: DeepSeek V4 Flash 0423
|
DeepSeek | $0.14 | $0.28 | $0.17 | — | — | 1.05M |
|
DeepSeek: DeepSeek V4 Flash 0731
|
DeepSeek | $0.22 | $0.66 | $0.33 | 72 | 0.9s | 1.05M |
|
DeepSeek: DeepSeek V4 Flash 0731 (batch)
|
DeepSeek | $0.11 | $0.33 | $0.17 | — | — | 1.05M |
|
DeepSeek: DeepSeek V4 Flash Vision Exp
|
DeepSeek | $0.22 | $0.66 | $0.33 | 97 | 0.7s | 1.05M |
|
DeepSeek: DeepSeek V4 Flash Vision Exp (batch)
|
DeepSeek | $0.11 | $0.33 | $0.17 | — | — | 1.05M |
|
DeepSeek: DeepSeek V4 Pro 0423
|
DeepSeek | $1.20 | $1.20 | $1.20 | — | — | 1.05M |
|
DeepSeek: DeepSeek V4 Pro 0813
|
DeepSeek | $1.32 | $3.96 | $1.98 | 50 | 1.1s | 1.05M |
|
DeepSeek: DeepSeek V4 Pro 0813 (batch)
|
DeepSeek | $0.66 | $1.98 | $0.99 | — | — | 1.05M |
|
DeepSeek: DeepSeek V4.1 Flash
|
DeepSeek | $0.22 | $0.66 | $0.33 | 78 | 1.1s | 1.05M |
Meta: Muse Glimmer 30B
|
Meta | $0.35 | $1.50 | $0.64 | 58 | 0.3s | 131k |
|
Meta: Muse Glimmer 30B (batch)
|
Meta | $0.17 | $0.75 | $0.32 | — | — | 131k |
MiniMax: MiniMax M2.7
|
MiniMax | $0.30 | $1.20 | $0.53 | — | — | 197k |
MoonshotAI: Kimi K2.6
|
Moonshot AI | $0.95 | $4.00 | $1.71 | 48 | 0.7s | 262k |
|
MoonshotAI: Kimi K2.7 Code
|
Moonshot AI | $0.95 | $4.00 | $1.71 | 46 | 0.9s | 262k |
|
MoonshotAI: Kimi K3
|
Moonshot AI | $3.00 | $15.00 | $6.00 | 84 | 0.8s | 1.05M |
OpenAI: gpt-oss-20b
|
OpenAI | $0.070 | $0.30 | $0.13 | — | — | 131k |
Qwen: Qwen3.8 2.4T A95B
|
Alibaba | $2.00 | $6.00 | $3.00 | 46 | 5.4s | 262k |
Z.ai: GLM 5.1
|
Z.ai | $1.40 | $4.40 | $2.15 | — | — | 203k |
|
Z.ai: GLM 5.2
|
Z.ai | $1.40 | $4.40 | $2.15 | 116 | 0.7s | 1.05M |
|
Z.ai: GLM 5.2 (batch)
|
Z.ai | $0.70 | $2.20 | $1.07 | — | — | 1.05M |
|
Z.ai: GLM 5.3
|
Z.ai | $1.40 | $4.40 | $2.15 | 54 | 1.6s | 1.05M |
|
Z.ai: GLM 5.3 (batch)
|
Z.ai | $0.70 | $2.20 | $1.07 | — | — | 1.05M |
|
Z.ai: GLM 5.3 Flash
|
Z.ai | $0.15 | $0.50 | $0.24 | 61 | 1.2s | 1.05M |
|
Z.ai: GLM 5.3 Flash (batch)
|
Z.ai | $0.075 | $0.25 | $0.12 | — | — | 1.05M |
The Known Good
Source. Prices, endpoint availability and throughput for Fireworks are ingested from OpenRouter;
model metadata is enriched from models.dev. A blank cell means the figure was not published for that endpoint —
it never means zero. Blended price is (3 × input + output) ÷ 4, the same 3:1 blend used everywhere on this
site; see methodology. Quality scores are not provider-specific:
a model scores the same wherever it is hosted, so quality lives on the
model pages.