Model time 1.1
- Models
- 21
- Median model time
- 40m 43s
- Total model time
- 19h 29m
- Total end-to-end time
- 136h 10m
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. Grok 4.7 (high): 20m 31s. 41s per round · 689 calls. View results for Grok 4.7 (high)
- 2. Gemini 3.7 Flash (high): 25m 46s. 52s per round · 525 calls. View results for Gemini 3.7 Flash (high)
- 3. Claude Sonnet 5.5 (high): 27m 34s. 55s per round · 562 calls. View results for Claude Sonnet 5.5 (high)
- 4. GPT-6 Sol (high): 32m 3s. 1m 4s per round · 571 calls. View results for GPT-6 Sol (high)
- 5. gpt-oss-120B (high): 32m 30s. 1m 5s per round · 662 calls. View results for gpt-oss-120B (high)
- 6. GPT-6 Luna (high): 35m 28s. 1m 11s per round · 659 calls. View results for GPT-6 Luna (high)
- 7. MiniMax M3 (high): 35m 46s. 1m 12s per round · 890 calls. View results for MiniMax M3 (high)
- 8. Claude Opus 5.5 (high): 38m 35s. 1m 17s per round · 479 calls. View results for Claude Opus 5.5 (high)
- 9. Claude Haiku 5.5 (high): 39m 28s. 1m 19s per round · 625 calls. View results for Claude Haiku 5.5 (high)
- 10. GPT-6.1 Sol (high): 40m 2s. 1m 20s per round · 486 calls. View results for GPT-6.1 Sol (high)
- 11. GPT-6 Astra (high): 40m 43s. 1m 21s per round · 440 calls. View results for GPT-6 Astra (high)
- 12. Claude Opus 5 (high): 41m 2s. 1m 22s per round · 570 calls. View results for Claude Opus 5 (high)
- 13. GPT-5.6 Luna (high): 41m 23s. 1m 23s per round · 658 calls. View results for GPT-5.6 Luna (high)
- 14. Gemini 3.8 Flash (high): 51m 12s. 1m 42s per round · 476 calls. View results for Gemini 3.8 Flash (high)
- 15. GLM-5.3-Flash (high): 51m 31s. 1m 43s per round · 720 calls. View results for GLM-5.3-Flash (high)
- 16. Muse Wonder (high): 53m 29s. 1m 47s per round · 618 calls. View results for Muse Wonder (high)
- 17. GPT-5.6 Sol (high): 57m 25s. 1m 55s per round · 571 calls. View results for GPT-5.6 Sol (high)
- 18. Gemini 4 Argon (high): 1h 6m. 2m 12s per round · 440 calls. View results for Gemini 4 Argon (high)
- 19. Claude Fable 5.1 (high): 1h 19m. 2m 39s per round · 481 calls. View results for Claude Fable 5.1 (high)
- 20. Muse Spark 1.3 Contributor (high): 2h 33m. 5m 6s per round · 580 calls. View results for Muse Spark 1.3 Contributor (high)
- 21. Grok 4.6 (high): 3h 25m. 6m 51s per round · 544 calls. View results for Grok 4.6 (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. GPT-6 Sol (high): 1h 37m. 3m 14s per round · 32m 3s model time. View results for GPT-6 Sol (high)
- 2. Gemini 3.7 Flash (high): 1h 53m. 3m 47s per round · 25m 46s model time. View results for Gemini 3.7 Flash (high)
- 3. GPT-6 Luna (high): 1h 54m. 3m 48s per round · 35m 28s model time. View results for GPT-6 Luna (high)
- 4. Claude Sonnet 5.5 (high): 1h 57m. 3m 55s per round · 27m 34s model time. View results for Claude Sonnet 5.5 (high)
- 5. GPT-6 Astra (high): 1h 58m. 3m 56s per round · 40m 43s model time. View results for GPT-6 Astra (high)
- 6. Gemini 4 Argon (high): 2h 18m. 4m 36s per round · 1h 6m model time. View results for Gemini 4 Argon (high)
- 7. Muse Wonder (high): 2h 24m. 4m 48s per round · 53m 29s model time. View results for Muse Wonder (high)
- 8. Claude Haiku 5.5 (high): 2h 40m. 5m 20s per round · 39m 28s model time. View results for Claude Haiku 5.5 (high)
- 9. gpt-oss-120B (high): 2h 53m. 5m 46s per round · 32m 30s model time. View results for gpt-oss-120B (high)
- 10. GLM-5.3-Flash (high): 2h 58m. 5m 56s per round · 51m 31s model time. View results for GLM-5.3-Flash (high)
- 11. Gemini 3.8 Flash (high): 2h 58m. 5m 56s per round · 51m 12s model time. View results for Gemini 3.8 Flash (high)
- 12. Claude Fable 5.1 (high): 3h 8m. 6m 17s per round · 1h 19m model time. View results for Claude Fable 5.1 (high)
- 13. Claude Opus 5.5 (high): 3h 51m. 7m 41s per round · 38m 35s model time. View results for Claude Opus 5.5 (high)
- 14. Grok 4.7 (high): 4h 41m. 9m 22s per round · 20m 31s model time. View results for Grok 4.7 (high)
- 15. Grok 4.6 (high): 7h 2m. 14m 3s per round · 3h 25m model time. View results for Grok 4.6 (high)
- 16. GPT-6.1 Sol (high): 8h 6m. 16m 12s per round · 40m 2s model time. View results for GPT-6.1 Sol (high)
- 17. MiniMax M3 (high): 8h 54m. 17m 48s per round · 35m 46s model time. View results for MiniMax M3 (high)
- 18. GPT-5.6 Sol (high): 9h 2m. 18m 4s per round · 57m 25s model time. View results for GPT-5.6 Sol (high)
- 19. GPT-5.6 Luna (high): 9h 38m. 19m 17s per round · 41m 23s model time. View results for GPT-5.6 Luna (high)
- 20. Claude Opus 5 (high): 20h 9m. 40m 18s per round · 41m 2s model time. View results for Claude Opus 5 (high)
- 21. Muse Spark 1.3 Contributor (high): 36h 7m. 1h 12m per round · 2h 33m model time. View results for Muse Spark 1.3 Contributor (high)
| Model-time rank | ||||||
|---|---|---|---|---|---|---|
| Rank 1 | Grok 4.7 (high) | 20m 31s | 4h 41m | 41s | 1.8s | 689 |
| Rank 2 | Gemini 3.7 Flash (high) | 25m 46s | 1h 53m | 52s | 2.9s | 525 |
| Rank 3 | Claude Sonnet 5.5 (high) | 27m 34s | 1h 57m | 55s | 2.9s | 562 |
| Rank 4 | GPT-6 Sol (high) | 32m 3s | 1h 37m | 1m 4s | 3.4s | 571 |
| Rank 5 | gpt-oss-120B (high) | 32m 30s | 2h 53m | 1m 5s | 2.9s | 662 |
| Rank 6 | GPT-6 Luna (high) | 35m 28s | 1h 54m | 1m 11s | 3.2s | 659 |
| Rank 7 | MiniMax M3 (high) | 35m 46s | 8h 54m | 1m 12s | 2.4s | 890 |
| Rank 8 | Claude Opus 5.5 (high) | 38m 35s | 3h 51m | 1m 17s | 4.8s | 479 |
| Rank 9 | Claude Haiku 5.5 (high) | 39m 28s | 2h 40m | 1m 19s | 3.8s | 625 |
| Rank 10 | GPT-6.1 Sol (high) | 40m 2s | 8h 6m | 1m 20s | 4.9s | 486 |
| Rank 11 | GPT-6 Astra (high) | 40m 43s | 1h 58m | 1m 21s | 5.6s | 440 |
| Rank 12 | Claude Opus 5 (high) | 41m 2s | 20h 9m | 1m 22s | 4.3s | 570 |
| Rank 13 | GPT-5.6 Luna (high) | 41m 23s | 9h 38m | 1m 23s | 3.8s | 658 |
| Rank 14 | Gemini 3.8 Flash (high) | 51m 12s | 2h 58m | 1m 42s | 6.5s | 476 |
| Rank 15 | GLM-5.3-Flash (high) | 51m 31s | 2h 58m | 1m 43s | 4.3s | 720 |
| Rank 16 | Muse Wonder (high) | 53m 29s | 2h 24m | 1m 47s | 5.2s | 618 |
| Rank 17 | GPT-5.6 Sol (high) | 57m 25s | 9h 2m | 1m 55s | 6.0s | 571 |
| Rank 18 | Gemini 4 Argon (high) | 1h 6m | 2h 18m | 2m 12s | 9.0s | 440 |
| Rank 19 | Claude Fable 5.1 (high) | 1h 19m | 3h 8m | 2m 39s | 9.9s | 481 |
| Rank 20 | Muse Spark 1.3 Contributor (high) | 2h 33m | 36h 7m | 5m 6s | 16s | 580 |
| Rank 21 | Grok 4.6 (high) | 3h 25m | 7h 2m | 6m 51s | 23s | 544 |
Rank 1Grok 4.7 (high)xai20m 31sModel time
- End-to-end time
- 4h 41m
- Model time / round
- 41s
Rank 2Gemini 3.7 Flash (high)google-ai-studio25m 46sModel time
- End-to-end time
- 1h 53m
- Model time / round
- 52s
Rank 3Claude Sonnet 5.5 (high)anthropic27m 34sModel time
- End-to-end time
- 1h 57m
- Model time / round
- 55s
Rank 4GPT-6 Sol (high)openai32m 3sModel time
- End-to-end time
- 1h 37m
- Model time / round
- 1m 4s
Rank 5gpt-oss-120B (high)cerebras32m 30sModel time
- End-to-end time
- 2h 53m
- Model time / round
- 1m 5s
Rank 6GPT-6 Luna (high)openai35m 28sModel time
- End-to-end time
- 1h 54m
- Model time / round
- 1m 11s
Rank 7MiniMax M3 (high)coreweave35m 46sModel time
- End-to-end time
- 8h 54m
- Model time / round
- 1m 12s
Rank 8Claude Opus 5.5 (high)anthropic38m 35sModel time
- End-to-end time
- 3h 51m
- Model time / round
- 1m 17s
Rank 9Claude Haiku 5.5 (high)anthropic39m 28sModel time
- End-to-end time
- 2h 40m
- Model time / round
- 1m 19s
Rank 10GPT-6.1 Sol (high)openai40m 2sModel time
- End-to-end time
- 8h 6m
- Model time / round
- 1m 20s
Rank 11GPT-6 Astra (high)openai40m 43sModel time
- End-to-end time
- 1h 58m
- Model time / round
- 1m 21s
Rank 12Claude Opus 5 (high)anthropic41m 2sModel time
- End-to-end time
- 20h 9m
- Model time / round
- 1m 22s
Rank 13GPT-5.6 Luna (high)openai41m 23sModel time
- End-to-end time
- 9h 38m
- Model time / round
- 1m 23s
Rank 14Gemini 3.8 Flash (high)google-ai-studio51m 12sModel time
- End-to-end time
- 2h 58m
- Model time / round
- 1m 42s
Rank 15GLM-5.3-Flash (high)z-ai51m 31sModel time
- End-to-end time
- 2h 58m
- Model time / round
- 1m 43s
Rank 16Muse Wonder (high)meta53m 29sModel time
- End-to-end time
- 2h 24m
- Model time / round
- 1m 47s
Rank 17GPT-5.6 Sol (high)openai57m 25sModel time
- End-to-end time
- 9h 2m
- Model time / round
- 1m 55s
Rank 18Gemini 4 Argon (high)google-vertex1h 6mModel time
- End-to-end time
- 2h 18m
- Model time / round
- 2m 12s
Rank 19Claude Fable 5.1 (high)anthropic1h 19mModel time
- End-to-end time
- 3h 8m
- Model time / round
- 2m 39s
Rank 20Muse Spark 1.3 Contributor (high)meta2h 33mModel time
- End-to-end time
- 36h 7m
- Model time / round
- 5m 6s
Rank 21Grok 4.6 (high)xai3h 25mModel time
- End-to-end time
- 7h 2m
- Model time / round
- 6m 51s
Models rank by model time. The end-to-end chart is only sorted by duration, because answer checks and rate limits dominate it.