LMArena Multi-turn arena · published July 20, 2026
Anthropic leads the LMArena multi-turn arena.
As of July 20, 2026, Anthropic leads the LMArena multi-turn arena with claude-opus-4-7 at 1518.8 across 9,076 battles, according to LMArena.
Current through July 20, 2026. 145 published snapshots. 377 of 381 models resolved (4 held for exact-match review).
| Rank | LMArena rating | Organization | Model | Reported interval | Gap to leader | Battles |
|---|---|---|---|---|---|---|
| 01 | 1518.8 | Anthropic | Claude Opus 4.7 | 1511.1 to 1526.4 | Leader | 9,076 |
| 02 | 1517.9 | Anthropic | claude-opus-4-6-thinking | 1510.9 to 1524.8 | -0.9 | 10,654 |
| 03 | 1517.5 | Anthropic | claude-opus-4-7-thinking | 1509.8 to 1525.2 | -1.3 | 8,703 |
| 04 | 1514.9 | Anthropic | Claude Fable 5 | 1499.7 to 1530.1 | -3.9 | 1,637 |
| 05 | 1510.6 | Anthropic | claude-opus-4-6 | 1503.9 to 1517.3 | -8.2 | 11,653 |
| 06 | 1506.3 | Anthropic | claude-opus-4-8-thinking | 1497.2 to 1515.4 | -12.5 | 5,367 |
| 07 | 1505.2 | Meta | muse-spark-1.1 | 1487.2 to 1523.2 | -13.6 | 1,199 |
| 08 | 1499.4 | Anthropic | claude-opus-4-8 | 1490.4 to 1508.4 | -19.4 | 5,652 |
| 09 | 1495.3 | Gemini 3 Pro | 1487.5 to 1503.2 | -23.5 | 6,766 | |
| 10 | 1494.4 | OpenAI | gpt-5.2-chat-latest-20260210 | 1486.4 to 1502.5 | -24.4 | 6,487 |
| 11 | 1493.2 | OpenAI | gpt-5.4-high | 1486.3 to 1500.2 | -25.6 | 10,720 |
| 12 | 1492.5 | gemini-3.1-pro-preview | 1486.2 to 1498.8 | -26.3 | 14,472 | |
| 13 | 1492.2 | Meta | Muse Spark | 1478.7 to 1505.7 | -26.6 | 2,164 |
| 14 | 1489.2 | OpenAI | gpt-5.5-high | 1481.1 to 1497.2 | -29.6 | 7,728 |
| 15 | 1488.6 | gemini-3.5-flash-high | 1474.1 to 1503.2 | -30.2 | 1,843 | |
| 16 | 1487.3 | Anthropic | claude-opus-4-5-20251101-thinking-32k | 1479.4 to 1495.1 | -31.5 | 6,468 |
| 17 | 1484.0 | Anthropic | claude-opus-4-5-20251101 | 1477.8 to 1490.2 | -34.8 | 12,676 |
| 18 | 1483.6 | Anthropic | claude-sonnet-4-6 | 1476.5 to 1490.6 | -35.2 | 10,182 |
| 19 | 1483.5 | gemini-3-flash | 1474.8 to 1492.1 | -35.3 | 5,320 | |
| 20 | 1483.4 | Z.ai | glm-5.1 | 1474.2 to 1492.6 | -35.4 | 4,945 |
| 21 | 1483.4 | Alibaba | qwen3.7-max-preview | 1459.8 to 1506.9 | -35.4 | 659 |
| 22 | 1482.7 | gemini-3.5-flash-medium | 1469.8 to 1495.6 | -36.1 | 2,317 | |
| 23 | 1482.4 | OpenAI | gpt-5.5-instant | 1473.0 to 1491.9 | -36.4 | 4,906 |
| 24 | 1482.0 | xAI | grok-4.20-beta1 | 1472.4 to 1491.6 | -36.8 | 4,384 |
| 25 | 1481.8 | OpenAI | gpt-5.5 | 1473.8 to 1489.7 | -37.0 | 8,151 |
Frontier race
Category Kings: Multi-turn arena frontier by organization
Anthropic holds the multi-turn arena frontier at 1519 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-opus-4-7 | 1518.8 | January 5, 2025 |
| Meta | meta/muse-spark-1.1 | 1505.2 | January 5, 2025 |
| google/gemini-3-pro | 1495.3 | January 5, 2025 | |
| OpenAI | openai/gpt-5.2-chat-latest-20260210 | 1494.4 | January 5, 2025 |
| Alibaba | alibaba/qwen3.7-max-preview | 1483.4 | January 5, 2025 |
| Z.ai | zai/glm-5.1 | 1483.4 | January 5, 2025 |
| xAI | xai/grok-4.20-beta1 | 1482.0 | January 5, 2025 |
| DeepSeek | deepseek/deepseek-v4-pro | 1472.8 | January 5, 2025 |
| Mistral AI | mistral/mistral-medium-3.5 | 1431.8 | January 5, 2025 |
Definitions
What these terms mean.
- Multi-turn arena
- The LMArena style-controlled battles with more than one conversational turn. It reflects preference across a dialogue, not long-horizon task completion.
- 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 Opus 4.7, Claude Fable 5, Gemini 3 Pro.
Multi-turn arena questions
Which model leads the Multi-turn arena right now?
According to the LMArena multi-turn conversation board published July 20, 2026, Anthropic leads with claude-opus-4-7 at 1518.8, across 9,076 battles. That is 0.9 points ahead of the next board entry inside this one snapshot.
Is this a Model Gauntlet multi-turn conversation ranking?
No. This board mirrors one named source, the LMArena multi-turn conversation 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 multi-turn conversation board?
The reviewed artifact covers 381 models across 61 organizations, with 377 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 multi-turn conversation gaps with caution.