LMArena Long queries arena · published July 20, 2026
Anthropic leads the LMArena long queries arena.
As of July 20, 2026, Anthropic leads the LMArena long queries arena with claude-opus-4-6-thinking at 1524.0 across 24,855 battles, according to LMArena.
Current through July 20, 2026. 145 published snapshots. 378 of 382 models resolved (4 held for exact-match review).
| Rank | LMArena rating | Organization | Model | Reported interval | Gap to leader | Battles |
|---|---|---|---|---|---|---|
| 01 | 1524.0 | Anthropic | claude-opus-4-6-thinking | 1518.5 to 1529.5 | Leader | 24,855 |
| 02 | 1524.0 | Anthropic | Claude Fable 5 | 1514.1 to 1533.8 | -0.0 | 4,362 |
| 03 | 1517.0 | Anthropic | claude-opus-4-7-thinking | 1511.1 to 1523.0 | -7.0 | 21,959 |
| 04 | 1517.0 | Anthropic | claude-opus-4-6 | 1511.7 to 1522.4 | -7.0 | 27,072 |
| 05 | 1508.8 | Anthropic | Claude Opus 4.7 | 1502.9 to 1514.7 | -15.2 | 22,653 |
| 06 | 1506.2 | Anthropic | claude-opus-4-8-thinking | 1499.2 to 1513.2 | -17.8 | 13,426 |
| 07 | 1500.4 | Anthropic | claude-opus-4-8 | 1493.4 to 1507.5 | -23.6 | 13,899 |
| 08 | 1497.8 | Alibaba | qwen3.7-max-preview | 1482.0 to 1513.6 | -26.2 | 1,597 |
| 09 | 1497.7 | gemini-3.1-pro-preview | 1492.6 to 1502.8 | -26.3 | 34,336 | |
| 10 | 1494.8 | Anthropic | claude-opus-4-5-20251101-thinking-32k | 1488.0 to 1501.6 | -29.2 | 9,235 |
| 11 | 1494.3 | Anthropic | claude-sonnet-4-6 | 1488.7 to 1499.9 | -29.7 | 23,226 |
| 12 | 1492.4 | Anthropic | claude-opus-4-5-20251101 | 1487.3 to 1497.6 | -31.6 | 22,656 |
| 13 | 1492.4 | Gemini 3 Pro | 1485.8 to 1499.0 | -31.6 | 10,480 | |
| 14 | 1489.1 | OpenAI | gpt-5.6-sol-xhigh | 1476.1 to 1502.1 | -34.9 | 2,324 |
| 15 | 1487.1 | Xiaomi | mimo-v2.5-pro | 1480.8 to 1493.4 | -36.9 | 17,664 |
| 16 | 1485.5 | Anthropic | claude-sonnet-4-5-20250929-thinking-32k | 1480.8 to 1490.2 | -38.5 | 24,489 |
| 17 | 1485.0 | OpenAI | gpt-5.5-high | 1478.9 to 1491.1 | -39.0 | 19,765 |
| 18 | 1484.9 | Anthropic | claude-opus-4-1-20250805-thinking-16k | 1478.7 to 1491.1 | -39.1 | 11,128 |
| 19 | 1484.1 | Z.ai | glm-5.1 | 1477.7 to 1490.6 | -39.9 | 13,101 |
| 20 | 1484.0 | Meta | muse-spark-1.1 | 1472.9 to 1495.1 | -40.0 | 3,145 |
| 21 | 1483.9 | OpenAI | gpt-5.4-high | 1478.3 to 1489.5 | -40.1 | 24,527 |
| 22 | 1483.4 | Anthropic | claude-sonnet-4-5-20250929 | 1478.6 to 1488.2 | -40.6 | 24,885 |
| 23 | 1482.0 | xAI | grok-4.5 | 1470.8 to 1493.2 | -42.0 | 3,138 |
| 24 | 1481.1 | Alibaba | qwen3.5-max-preview | 1473.7 to 1488.5 | -42.9 | 8,259 |
| 25 | 1481.0 | Z.ai | glm-5.2 (max) | 1473.0 to 1489.0 | -43.0 | 7,693 |
Frontier race
Category Kings: Long queries arena frontier by organization
Anthropic holds the long queries arena frontier at 1524 as of July 20, 2026.
Method: Frontier is the maximum resolved LMArena rating per organization on each publish date. Organizations limited to the union of the seven tracked homepage providers and organizations in the latest board top 25 by officialRank. Quarantined dates are excluded, and this category series keeps its own method epoch and is never joined to the legacy archive.
| Organization | Latest model | Latest LMArena rating | First tracked date |
|---|---|---|---|
| Anthropic | anthropic/claude-fable-5 | 1524.0 | January 5, 2025 |
| Alibaba | alibaba/qwen3.7-max-preview | 1497.8 | January 5, 2025 |
| google/gemini-3.1-pro-preview | 1497.7 | January 5, 2025 | |
| OpenAI | openai/gpt-5.6-sol-xhigh | 1489.1 | January 5, 2025 |
| Xiaomi | xiaomi/mimo-v2.5-pro | 1487.1 | December 23, 2025 |
| Z.ai | zai/glm-5.1 | 1484.1 | January 5, 2025 |
| Meta | meta/muse-spark-1.1 | 1484.0 | January 5, 2025 |
| xAI | xai/grok-4.5 | 1482.0 | January 5, 2025 |
| DeepSeek | deepseek/deepseek-v4-pro | 1473.2 | January 5, 2025 |
| Mistral AI | mistral/mistral-medium-3.5 | 1431.1 | January 5, 2025 |
Definitions
What these terms mean.
- Long queries arena
- The LMArena style-controlled text battles filtered to longer prompts as classified by the source. Prompt length is the source's cutoff, not a context-window benchmark.
- Frontier by organization
- The highest resolved score for each organization on each publish date. It tracks how far each lab has pushed this category, not an average of all its models.
- Confidence interval
- The reported 95% range LMArena publishes with each rating. Overlapping intervals mean this snapshot does not separate those ranks decisively.
Follow the evidence
Compare this board against the overall LMArena rankings, shortlist candidates with the Reasoning and research guide,read the Anthropic lab file, and see how rankings are selected. Model profiles on this board: Claude Fable 5, Claude Opus 4.7, Gemini 3 Pro.
Long queries arena questions
Which model leads the Long queries arena right now?
According to the LMArena long queries board published July 20, 2026, Anthropic leads with claude-opus-4-6-thinking at 1524.0, across 24,855 battles. That is 0.0 points ahead of the next board entry inside this one snapshot.
Is this a Model Gauntlet long queries ranking?
No. This board mirrors one named source, the LMArena long queries arena, published July 20, 2026. Model Gauntlet publishes no composite or consensus score and runs no evaluation of its own.
How many models and organizations are in the long queries board?
The reviewed artifact covers 382 models across 61 organizations, with 378 models resolved to exact identities and 4 held for review. The history spans 145 snapshots from January 5, 2025 to July 20, 2026.
Why do some scores have overlapping confidence intervals?
LMArena publishes a 95% confidence interval with each rating. Where two intervals overlap, this single snapshot does not separate those ranks decisively, so treat small long queries gaps with caution.