Provider Snapshot
2
59.25
910.00
Mar 8, 2026
Key Takeaways
2 fireworks models are actively benchmarked with 423 total measurements across 418 benchmark runs.
mixtral-8x22b leads the fleet with 67.30 tokens/second, while llama-3.3-70b delivers 51.20 tok/s.
Performance varies by 31.4% across the fireworks model lineup, indicating diverse optimization strategies for different use cases.
Avg time to first token across the fleet is 910.00 ms, showing good responsiveness for interactive applications.
Fastest Models
| Provider | Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|---|
| fireworks | mixtral-8x22b | 67.30 | 29.90 | 112.00 | 430.00 |
| fireworks | llama-3.3-70b | 51.20 | 2.64 | 94.60 | 1390.00 |
All Models
Complete list of all fireworks models tracked in the benchmark system. Click any model name to view detailed performance data.
| Provider | Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|---|
| fireworks | llama-3.3-70b | 51.20 | 2.64 | 94.60 | 1390.00 |
| fireworks | mixtral-8x22b | 67.30 | 29.90 | 112.00 | 430.00 |
Featured Models
Frequently Asked Questions
Based on recent tests, mixtral-8x22b shows the highest average throughput among tracked fireworks models.
This provider summary aggregates 423 individual prompts measured across 418 monitoring runs over the past month.