Best LLMs Ranked (October 2026): LMArena Top 30
Top 30 large language models by LMArena Text Arena score, with votes, license, API price and context window.
Source: LMArena Text Arena leaderboard (overall), updated Oct 20, 2026 · Data checked on 2026-10-21 · Machine-readable JSON
| Rank | Model | Org | Arena score | Votes | Price /1M (in/out) | Context | License |
|---|---|---|---|---|---|---|---|
| 1 | Gemini 4 Argon (High) | 1524 ±9 (prelim.) | 4,900 | $2 / $10 | 1M | Proprietary | |
| 2 | Muse Wonder (Max) | Meta | 1510 ±9 | 5,099 | $4 / $20 | 1M | Proprietary |
| 3 | Claude Opus 5.5 (High) | Anthropic | 1509 ±7 | 9,496 | $4 / $20 | 1M | Proprietary |
| 4 | Claude Fable 5 (High) | Anthropic | 1504 ±4 | 43,399 | $10 / $50 | 1M | Proprietary |
| 5 | Claude Opus 4.6 (High) | Anthropic | 1503 ±3 | 82,619 | $5 / $25 | 1M | Proprietary |
| 6 | Claude Fable 5.1 (Max) | Anthropic | 1503 ±6 | 15,623 | $10 / $50 | 1M | Proprietary |
| 7 | Claude Opus 4.7 (High) | Anthropic | 1501 ±4 | 65,131 | $5 / $25 | 1M | Proprietary |
| 8 | Claude Opus 4.6 | Anthropic | 1497 ±3 | 87,372 | $5 / $25 | 1M | Proprietary |
| 9 | Gemini 3.8 Flash (High) | 1496 ±5 (prelim.) | 41,423 | $0.75 / $3.75 | 1M | Proprietary | |
| 10 | Muse Spark 1.3 (Max) | Meta | 1495 ±5 | 19,323 | $1.25 / $4.25 | 1M | Proprietary |
| 11 | Claude Opus 4.7 | Anthropic | 1494 ±4 | 66,185 | $5 / $25 | 1M | Proprietary |
| 12 | Muse Spark 1.2 (xHigh) | Meta | 1492 ±7 | 9,578 | $1.25 / $4.25 | – | Proprietary |
| 13 | Claude Opus 5 (Max) | Anthropic | 1491 ±5 | 33,683 | $5 / $25 | 1M | Proprietary |
| 14 | Muse Spark 1.1 | Meta | 1490 ±4 | 44,815 | $1.25 / $4.25 | – | Proprietary |
| 15 | Claude Opus 5 (High) | Anthropic | 1490 ±4 | 69,307 | $5 / $25 | 1M | Proprietary |
| 16 | Muse Spark | Meta | 1489 ±6 | 14,153 | – | – | Proprietary |
| 17 | Gemini 3.7 Flash (High) | 1489 ±5 (prelim.) | 29,177 | $0.75 / $3.75 | 1M | Proprietary | |
| 18 | Kimi K3 (Max) | Moonshot | 1488 ±5 | 34,325 | – | – | Kimi K3 license |
| 19 | Gemini 3.1 Pro Preview | 1487 ±3 | 129,523 | $1 / $6 | 1M | Proprietary | |
| 20 | Gemini 3 Pro | 1485 ±4 | 42,059 | $2 / $12 | 1M | Proprietary | |
| 21 | GPT-5.6 Sol (xHigh) | OpenAI | 1484 ±4 | 41,803 | $4 / $20 | – | Proprietary |
| 22 | Gemini 3.6 Flash (High) | 1483 ±4 | 36,079 | $0.75 / $3.75 | 1M | Proprietary | |
| 23 | Qwen3.8-Max | Alibaba | 1483 ±5 | 29,418 | $2 / $6 | 1M | Proprietary |
| 24 | GPT-5.5 (High) | OpenAI | 1481 ±4 | 69,340 | $5 / $30 | 1.1M | Proprietary |
| 25 | Claude Opus 4.8 (High) | Anthropic | 1481 ±4 | 64,063 | $5 / $25 | 1M | Proprietary |
| 26 | GLM-5.3 (Max) | Z.ai | 1481 ±5 | 24,035 | $1.40 / $4.40 | 1M | MIT |
| 27 | MiMo-V2.6-Pro | Xiaomi | 1480 ±9 | 4,065 | $0.43 / $0.87 | 1.1M | MIT |
| 28 | GPT-6.1 Sol (Max) | OpenAI | 1479 ±8 | 7,453 | $2 / $10 | 1.1M | Proprietary |
| 29 | Gemini 3.5 Flash (High) | 1477 ±4 | 48,479 | $0.75 / $4.50 | 1M | Proprietary | |
| 30 | Claude Sonnet 5.5 (xHigh) | Anthropic | 1477 ±7 | 8,126 | $2 / $10 | 1M | Proprietary |
Methodology
Scores are LMArena Elo-style ratings from blind, crowd-sourced head-to-head votes (overall text category). ±N is LMArena's 95% confidence interval; models whose intervals overlap are statistically tied. "Preliminary" means few votes so far. Price ($ input / $ output per 1M tokens) and context are as shown on LMArena. We did not run our own benchmarks.
Get the 5 best new AI tools every week
One short email every Friday. No spam, unsubscribe anytime.