Gemini 4 Argon (High)
1,518±18
1,499CI1,536
aiqbench · LLM benchmarks
Benchmarks
- Sources
- 8
- Entries tracked
- 22,318
- Last sync
- 36 min36 minutes ago
All sources syncing normally
- Arena · Text· 4 hours ago
- Arena · WebDev· 4 hours ago
- Arena · Vision· 4 hours ago
- Arena · Agent· 4 hours ago
- GPQA Diamond· 36 minutes ago
- SWE-bench· 36 minutes ago
- Terminal-Bench· 36 minutes ago
- ARC-AGI-2· 36 minutes ago
Filter by org:
Released:
Claude Opus 5.5 (High)
1,498±14
1,485CI1,512
Claude Fable 5 (High)
1,493±7
1,486CI1,500
One grid, every column. Pick axes in the Axes panel — one axis ranks like a single board, several blend into a weighted composite, all of them read as the index. Rank is assigned before any filter and never renumbers.
| Model | Org | ± CI | Swing | Add to compare | |||||
|---|---|---|---|---|---|---|---|---|---|
| 1 | Gemini 4 Argon (High) | Oct 2026 | $2 / $10 | 1,518 | 1,499–1,536 | #1–10 | |||
| 2 | Claude Opus 5.5 (High) | anthropic | Oct 2026 | $4 / $20 | 1,498 | 1,485–1,512 | #2–11 | ||
| 3 | Claude Fable 5 (High) | anthropic | Sep 2026 | $10 / $50 | 1,493 | 1,486–1,500 | #4–21 | ||
| 4 | Claude Opus 4.6 (High) | anthropic | Sep 2026 | $5 / $25 | 1,491 | 1,485–1,497 | #2–25 | ||
| 5 | Claude Opus 4.7 (High) | anthropic | Sep 2026 | $5 / $25 | 1,481 | 1,474–1,487 | #1–18 | ||
| 6 | Gemini 3.7 Flash (High) | Oct 2026 | $0.75 / $3.75 | 1,480 | 1,472–1,488 | #9–57 | |||
| 7 | Gemini 3.8 Flash (High) | Oct 2026 | $0.83 / $4.13 | 1,477 | 1,470–1,484 | #2–30 | |||
| 8 | Claude Fable 5.1 (Max) | anthropic | Sep 2026 | $10 / $50 | 1,476 | 1,466–1,487 | #2–22 | ||
| 9 | Muse Wonder (Max) | meta | Oct 2026 | — | 1,474 | 1,453–1,494 | #14–56 | ||
| 10 | gpt-5.6-sol-xhigh | openai | Sep 2026 | $4 / $20 | 1,471 | 1,464–1,478 | #6–24 | ||
| 11 | Gemini 3.1 Pro Preview | Sep 2026 | $2 / $12 | 1,470 | 1,465–1,475 | #9–36 | |||
| 12 | Gemini 3 Pro | Jan 2026 | — | 1,469 | 1,462–1,477 | #23–91 | |||
| 13 | Claude Opus 5 (High) | anthropic | Sep 2026 | $5 / $25 | 1,462 | 1,456–1,468 | #5–11 | ||
| 14 | Muse Spark | meta | Sep 2026 | — | 1,462 | 1,450–1,474 | #16–41 | ||
| 15 | Kimi K3 (Max) | moonshot | Jun 2026 | $0.24 / $11.94 | 1,461 | 1,454–1,469 | #10–40 | ||
| 16 | gemini-3.6-flash-high | Sep 2026 | $0.75 / $3.75 | 1,461 | 1,453–1,468 | #16–95 | |||
| 17 | claude-opus-4-8-high | anthropic | Sep 2026 | $5 / $25 | 1,460 | 1,454–1,466 | #15–33 | ||
| 18 | claude-sonnet-5.5-xhigh | anthropic | Oct 2026 | $2 / $10 | 1,460 | 1,445–1,475 | #4–26 | ||
| 19 | gpt-6-astra-max | openai | Sep 2026 | $10 / $50 | 1,460 | 1,448–1,471 | #1–18 | ||
| 20 | gpt-6.1-sol-max | openai | Oct 2026 | $2 / $10 | 1,459 | 1,443–1,476 | #1–15 | ||
| 21 | qwen3.8-max | alibaba | Oct 2026 | $2 / $6 | 1,459 | 1,451–1,467 | #12–41 | ||
| 22 | Muse Spark 1.3 (Max) | meta | Sep 2026 | $1.25 / $4.25 | 1,458 | 1,448–1,468 | #13–33 | ||
| 23 | gemini-3.5-flash-medium | Sep 2026 | $1.50 / $9 | 1,457 | 1,451–1,464 | #10–26 | |||
| 24 | claude-opus-4-5-20251101-high-32k | anthropic | Nov 2025 | — | 1,456 | 1,449–1,464 | #36–94 | ||
| 25 | claude-opus-4-5-20251101 | anthropic | Nov 2025 | $5 / $25 | 1,454 | 1,449–1,460 | #19–62 | ||
| 26 | gemini-3-flash | Jan 2026 | — | 1,454 | 1,445–1,462 | #38–153 | |||
| 27 | Muse Spark 1.2 (xHigh) | meta | Oct 2026 | $1.25 / $4.25 | 1,453 | 1,440–1,467 | #13–112 | ||
| 28 | gpt-5.5-instant | openai | Sep 2026 | — | 1,451 | 1,443–1,460 | #52–114 | ||
| 29 | deepseek-v4.1-flash-max | deepseek | Oct 2026 | $0.30 / $1.20 | 1,450 | 1,439–1,461 | #14–30 | ||
| 30 | grok-4.20-beta1 | xai | Oct 2026 | — | 1,449 | 1,440–1,457 | #46–119 | ||
| 31 | gpt-5.5-high | openai | Sep 2026 | $5 / $30 | 1,449 | 1,443–1,455 | #9–34 | ||
| 32 | Muse Spark 1.1 | meta | Sep 2026 | $1.25 / $4.25 | 1,449 | 1,442–1,455 | #21–115 | ||
| 33 | glm-5.2-max | zai | Jun 2026 | $0.62 / $2.42 | 1,447 | 1,441–1,454 | #9–49 | ||
| 34 | mimo-v2.6-pro | xiaomi | Sep 2026 | $0.43 / $0.87 | 1,447 | 1,428–1,467 | #19–64 | ||
| 35 | grok-4.5 | xai | Sep 2026 | $2 / $6 | 1,446 | 1,439–1,453 | #29–70 |
Showing Entertainment & media exactly as the source published it — its own score and confidence interval, no blending. Select a second axis to build a weighted composite.