Model time 1.0
- Models
- 18
- Median model time
- 1h 11m
- Total model time
- 33h 50m
- Total end-to-end time
- 79h 0m
Model time
Model time across the run.
Each bar is the total time of all calls to the tested model, as reported by the provider. Shorter is faster. This is not the total benchmark runtime.
- 1. Llama 4 Maverick (non-thinking): 16m 16s. 28s per round · 1,132 calls. View results for Llama 4 Maverick (non-thinking)
- 2. Grok 4.5 (high): 40m 43s. 1m 10s per round · 564 calls. View results for Grok 4.5 (high)
- 3. Claude Opus 5 (high): 42m 27s. 1m 13s per round · 467 calls. View results for Claude Opus 5 (high)
- 4. Gemini 3.7 Flash (high): 44m 51s. 1m 17s per round · 525 calls. View results for Gemini 3.7 Flash (high)
- 5. GPT-5.6 Luna (high): 45m 12s. 1m 17s per round · 653 calls. View results for GPT-5.6 Luna (high)
- 6. Claude Sonnet 5 (high): 49m 5s. 1m 24s per round · 551 calls. View results for Claude Sonnet 5 (high)
- 7. gpt-oss-120B (high): 53m 48s. 1m 32s per round · 490 calls. View results for gpt-oss-120B (high)
- 8. Gemini 3.6 Flash (high): 1h 9m. 1m 59s per round · 625 calls. View results for Gemini 3.6 Flash (high)
- 9. Claude Fable 5 (high): 1h 9m. 1m 59s per round · 457 calls. View results for Claude Fable 5 (high)
- 10. Gemini 3.8 Flash (high): 1h 14m. 2m 6s per round · 495 calls. View results for Gemini 3.8 Flash (high)
- 11. GPT-5.6 Sol (high): 1h 19m. 2m 16s per round · 539 calls. View results for GPT-5.6 Sol (high)
- 12. GPT-6 Astra (high): 1h 35m. 2m 42s per round · 561 calls. View results for GPT-6 Astra (high)
- 13. Grok 4.6 (high): 1h 49m. 3m 8s per round · 535 calls. View results for Grok 4.6 (high)
- 14. Claude Fable 5.1 (high): 1h 55m. 3m 17s per round · 458 calls. View results for Claude Fable 5.1 (high)
- 15. Ox Alpha (high): 3h 3m. 5m 14s per round · 648 calls. View results for Ox Alpha (high)
- 16. GPT-5 Nano (medium): 3h 14m. 5m 32s per round · 575 calls. View results for GPT-5 Nano (medium)
- 17. Kimi K3 (high): 4h 2m. 6m 54s per round · 479 calls. View results for Kimi K3 (high)
- 18. Mistral Medium 3.5 (high): 8h 29m. 14m 32s per round · 733 calls. View results for Mistral Medium 3.5 (high)
End-to-end time
End-to-end benchmark time.
Each bar is the clock time from start to finish of the run. It includes model calls, answer checks, web search, and waiting for provider rate limits, so it says more about the test setup than about the model. It is sorted, not ranked.
- 1. Llama 4 Maverick (non-thinking): 1h 53m. 3m 14s per round · 16m 16s model time. View results for Llama 4 Maverick (non-thinking)
- 2. Claude Opus 5 (high): 2h 4m. 3m 33s per round · 42m 27s model time. View results for Claude Opus 5 (high)
- 3. gpt-oss-120B (high): 2h 13m. 3m 49s per round · 53m 48s model time. View results for gpt-oss-120B (high)
- 4. Grok 4.5 (high): 2h 15m. 3m 51s per round · 40m 43s model time. View results for Grok 4.5 (high)
- 5. Claude Sonnet 5 (high): 2h 16m. 3m 52s per round · 49m 5s model time. View results for Claude Sonnet 5 (high)
- 6. Claude Fable 5 (high): 2h 40m. 4m 34s per round · 1h 9m model time. View results for Claude Fable 5 (high)
- 7. Gemini 3.7 Flash (high): 2h 40m. 4m 35s per round · 44m 51s model time. View results for Gemini 3.7 Flash (high)
- 8. GPT-5.6 Luna (high): 2h 47m. 4m 47s per round · 45m 12s model time. View results for GPT-5.6 Luna (high)
- 9. Gemini 3.6 Flash (high): 2h 52m. 4m 55s per round · 1h 9m model time. View results for Gemini 3.6 Flash (high)
- 10. Gemini 3.8 Flash (high): 2h 53m. 4m 56s per round · 1h 14m model time. View results for Gemini 3.8 Flash (high)
- 11. GPT-5.6 Sol (high): 2h 57m. 5m 3s per round · 1h 19m model time. View results for GPT-5.6 Sol (high)
- 12. Claude Fable 5.1 (high): 3h 24m. 5m 50s per round · 1h 55m model time. View results for Claude Fable 5.1 (high)
- 13. Grok 4.6 (high): 4h 40m. 8m 0s per round · 1h 49m model time. View results for Grok 4.6 (high)
- 14. GPT-5 Nano (medium): 4h 55m. 8m 26s per round · 3h 14m model time. View results for GPT-5 Nano (medium)
- 15. Kimi K3 (high): 5h 17m. 9m 3s per round · 4h 2m model time. View results for Kimi K3 (high)
- 16. Ox Alpha (high): 7h 33m. 12m 57s per round · 3h 3m model time. View results for Ox Alpha (high)
- 17. GPT-6 Astra (high): 10h 41m. 18m 19s per round · 1h 35m model time. View results for GPT-6 Astra (high)
- 18. Mistral Medium 3.5 (high): 14h 58m. 25m 40s per round · 8h 29m model time. View results for Mistral Medium 3.5 (high)
| Model-time rank | ||||||
|---|---|---|---|---|---|---|
| Rank 1 | Llama 4 Maverick (non-thinking) | 16m 16s | 1h 53m | 28s | 0.9s | 1,132 |
| Rank 2 | Grok 4.5 (high) | 40m 43s | 2h 15m | 1m 10s | 4.3s | 564 |
| Rank 3 | Claude Opus 5 (high) | 42m 27s | 2h 4m | 1m 13s | 5.5s | 467 |
| Rank 4 | Gemini 3.7 Flash (high) | 44m 51s | 2h 40m | 1m 17s | 5.1s | 525 |
| Rank 5 | GPT-5.6 Luna (high) | 45m 12s | 2h 47m | 1m 17s | 4.2s | 653 |
| Rank 6 | Claude Sonnet 5 (high) | 49m 5s | 2h 16m | 1m 24s | 5.3s | 551 |
| Rank 7 | gpt-oss-120B (high) | 53m 48s | 2h 13m | 1m 32s | 6.6s | 490 |
| Rank 8 | Gemini 3.6 Flash (high) | 1h 9m | 2h 52m | 1m 59s | 6.6s | 625 |
| Rank 9 | Claude Fable 5 (high) | 1h 9m | 2h 40m | 1m 59s | 9.1s | 457 |
| Rank 10 | Gemini 3.8 Flash (high) | 1h 14m | 2h 53m | 2m 6s | 8.9s | 495 |
| Rank 11 | GPT-5.6 Sol (high) | 1h 19m | 2h 57m | 2m 16s | 8.8s | 539 |
| Rank 12 | GPT-6 Astra (high) | 1h 35m | 10h 41m | 2m 42s | 10s | 561 |
| Rank 13 | Grok 4.6 (high) | 1h 49m | 4h 40m | 3m 8s | 12s | 535 |
| Rank 14 | Claude Fable 5.1 (high) | 1h 55m | 3h 24m | 3m 17s | 15s | 458 |
| Rank 15 | Ox Alpha (high) | 3h 3m | 7h 33m | 5m 14s | 17s | 648 |
| Rank 16 | GPT-5 Nano (medium) | 3h 14m | 4h 55m | 5m 32s | 20s | 575 |
| Rank 17 | Kimi K3 (high) | 4h 2m | 5h 17m | 6m 54s | 30s | 479 |
| Rank 18 | Mistral Medium 3.5 (high) | 8h 29m | 14h 58m | 14m 32s | 42s | 733 |
Rank 1Llama 4 Maverick (non-thinking)parasail16m 16sModel time
- End-to-end time
- 1h 53m
- Model time / round
- 28s
Rank 2Grok 4.5 (high)xai40m 43sModel time
- End-to-end time
- 2h 15m
- Model time / round
- 1m 10s
Rank 3Claude Opus 5 (high)anthropic42m 27sModel time
- End-to-end time
- 2h 4m
- Model time / round
- 1m 13s
Rank 4Gemini 3.7 Flash (high)google-ai-studio44m 51sModel time
- End-to-end time
- 2h 40m
- Model time / round
- 1m 17s
Rank 5GPT-5.6 Luna (high)openai45m 12sModel time
- End-to-end time
- 2h 47m
- Model time / round
- 1m 17s
Rank 6Claude Sonnet 5 (high)anthropic49m 5sModel time
- End-to-end time
- 2h 16m
- Model time / round
- 1m 24s
Rank 7gpt-oss-120B (high)cerebras53m 48sModel time
- End-to-end time
- 2h 13m
- Model time / round
- 1m 32s
Rank 8Gemini 3.6 Flash (high)google-vertex1h 9mModel time
- End-to-end time
- 2h 52m
- Model time / round
- 1m 59s
Rank 9Claude Fable 5 (high)anthropic1h 9mModel time
- End-to-end time
- 2h 40m
- Model time / round
- 1m 59s
Rank 10Gemini 3.8 Flash (high)google-ai-studio1h 14mModel time
- End-to-end time
- 2h 53m
- Model time / round
- 2m 6s
Rank 11GPT-5.6 Sol (high)openai1h 19mModel time
- End-to-end time
- 2h 57m
- Model time / round
- 2m 16s
Rank 12GPT-6 Astra (high)openai1h 35mModel time
- End-to-end time
- 10h 41m
- Model time / round
- 2m 42s
Rank 13Grok 4.6 (high)xai1h 49mModel time
- End-to-end time
- 4h 40m
- Model time / round
- 3m 8s
Rank 14Claude Fable 5.1 (high)anthropic1h 55mModel time
- End-to-end time
- 3h 24m
- Model time / round
- 3m 17s
Rank 15Ox Alpha (high)stealth3h 3mModel time
- End-to-end time
- 7h 33m
- Model time / round
- 5m 14s
Rank 16GPT-5 Nano (medium)openai3h 14mModel time
- End-to-end time
- 4h 55m
- Model time / round
- 5m 32s
Rank 17Kimi K3 (high)moonshotai4h 2mModel time
- End-to-end time
- 5h 17m
- Model time / round
- 6m 54s
Rank 18Mistral Medium 3.5 (high)mistral8h 29mModel time
- End-to-end time
- 14h 58m
- Model time / round
- 14m 32s
Models rank by model time. The end-to-end chart is only sorted by duration, because answer checks and rate limits dominate it.