Holding context and coherence across a long back-and-forth. Ranked by Arena text / multi_turn, priced against live API rates. 45 models have a published score on this board.
Capability published 2026-10-22 · pricing retrieved 2026-10-25
google/gemini-4-argongoogle/gemini-4-argondeepseek/deepseek-v4-flash-0731google/gemini-4-argonRows marked frontier are Pareto-optimal — nothing we track is both stronger and cheaper. "Strength" rescales the board so 100% = parity with the leader.
| # | Model | Provider | Score | Strength | $/1M | Context | Tools |
|---|---|---|---|---|---|---|---|
| 1 | Gemini 4 Argongoogle/gemini-4-argon | 1547 | 100% | $4.00 | 1.048576M | tools | |
| 2 | Claude Fable 5anthropic/claude-fable-5 | Anthropic | 1504 | 88% | $20.00 | 1M | tools |
| 3 | Claude Opus 5.5anthropic/claude-opus-5.5 | Anthropic | 1504 | 88% | $8.00 | 1M | tools |
| 4 | Gemini 3.8 Flashgoogle/gemini-3.8-flash | 1503 | 88% | $1.66 | 1.048576M | tools | |
| 5 | Claude Opus 5anthropic/claude-opus-5 | Anthropic | 1500 | 87% | $10.00 | 1M | tools |
| 6 | Claude Fable 5.1anthropic/claude-fable-5.1 | Anthropic | 1495 | 85% | $20.00 | 1M | tools |
| 7 | Qwen3.8 Max (0902)qwen/qwen3.8-max-0902 | Qwen (Alibaba) | 1490 | 84% | $3.00 | 1M | tools |
| 8 | Kimi K3moonshotai/kimi-k3 | Moonshot AI | 1489 | 84% | $2.88 | 1.048576M | tools |
| 9 | Gemini 3.7 Flashgoogle/gemini-3.7-flash | 1487 | 83% | $1.50 | 1.048576M | tools | |
| 10 | Gemini 3.6 Flashgoogle/gemini-3.6-flash | 1481 | 81% | $1.50 | 1.048576M | tools | |
| 11 | Claude Opus 4.8anthropic/claude-opus-4.8 | Anthropic | 1476 | 80% | $10.00 | 1M | tools |
| 12 | GLM 5.1z-ai/glm-5.1 | Z.ai (Zhipu) | 1472 | 79% | $2.15 | 205K | tools |
| 13 | GLM 5.3 Flashz-ai/glm-5.3-flash | Z.ai (Zhipu) | 1471 | 79% | $0.24 | 1.048576M | tools |
| 14 | GLM 5.3z-ai/glm-5.3 | Z.ai (Zhipu) | 1470 | 78% | $0.41 | 1.048576M | tools |
| 15 | GLM 5.2z-ai/glm-5.2 | Z.ai (Zhipu) | 1469 | 78% | $1.77 | 1.048576M | tools |
| 16 | DeepSeek V4 Pro 0813deepseek/deepseek-v4-pro-0813 | DeepSeek | 1468 | 78% | $0.99 | 1.048576M | tools |
| 17 | DeepSeek V4 Pro 0423deepseek/deepseek-v4-pro | DeepSeek | 1468 | 78% | $0.26 | 1.048576M | tools |
| 18 | Claude Sonnet 5.5anthropic/claude-sonnet-5.5 | Anthropic | 1465 | 77% | $4.00 | 1M | tools |
| 19 | DeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash | DeepSeek | 1460 | 76% | $0.27 | 1.048576M | tools |
| 20 | Qwen3.7 Plusqwen/qwen3.7-plus | Qwen (Alibaba) | 1459 | 75% | $0.56 | 1M | tools |
| 21 | Grok 4.5x-ai/grok-4.5 | xAI | 1456 | 75% | $3.00 | 500K | tools |
| 22 | Grok 4.20x-ai/grok-4.20 | xAI | 1456 | 74% | $1.56 | 2M | tools |
| 23 | Claude Sonnet 5anthropic/claude-sonnet-5 | Anthropic | 1453 | 74% | $4.00 | 1M | tools |
| 24 | Kimi K2.6moonshotai/kimi-k2.6 | Moonshot AI | 1453 | 74% | $1.34 | 262K | tools |
| 25 | Grok 4.20 Multi-Agentx-ai/grok-4.20-multi-agent | xAI | 1452 | 73% | $1.56 | 2M | — |
Capability: Arena leaderboard dataset (CC-BY-4.0), Arena text / multi_turn, published 2026-10-22. Pricing: OpenRouter, retrieved 2026-10-25. (input x 3 + output x 1) / 4.
curl -s https://agentleaderboards.com/api/v1/route/long-conversation.json | jq .picksNo key, open CORS, refreshed daily. Set your own budget and constraints →
Writing & editing code · Autonomous agent work · Tool-call reliability · Following operator instructions · Recovering from shell errors · Hard reasoning · Mathematics · Precise instruction following · Long inputs · Creative writing · General assistant · Software & IT services · Legal & government · Medicine & healthcare · Science · Business & finance · Writing & language