Kimi K2.5 (Thinking)
Moonshot AIOpenModified MIT
Kimi K2.5 (Thinking) is a vision-language model from Moonshot AI. It ranks 46 of 167 on LMArena’s overall vision board with 1269±6 from 28,249 votes. Its best category is Entity recognition (#7 of 51). License: Modified MIT.
Key figures
- Developer
- Moonshot AI
- License
- Modified MIT
- First seen (our snapshots)
- 2026-10-26
Who it suits
- Overall #46 of 167: upper half.
- Strongest category: Entity recognition (#7 of 51).
- Open weights (Modified MIT): you can self-host it.
Neighbours on the overall board
| # | Model | Score | Votes | License | Δ vs this model |
|---|---|---|---|---|---|
| 43 |  GPT-5.6 Terra (xHigh) OpenAI | 1271 | 11,842 | Proprietary | +2 |
| 44 |  Qwen3.8-27B Alibaba | 1269 | 7,614 | Apache 2.0 | +0 |
| 45 |  Gemini 3.5 Flash-Lite Google | 1269 | 9,238 | Proprietary | +0 |
| 46 | Kimi K2.5 (Thinking) Moonshot AI | 1269 | 28,249 | Modified MIT | |
| 47 | GPT-5.2 Chat (20260210) OpenAI | 1268 | 16,465 | Proprietary | -1 |
| 48 | Gemini 3 Flash (Minimal) Google | 1266 | 37,173 | Proprietary | -3 |
| 49 | SStep 5 Preview StepFun | 1265 | 1,649 | Proprietary | -4 |
Alternatives · More from Moonshot AI
| # | Model | Score | Votes | License |
|---|---|---|---|---|
| 34 | Kimi K2.6 Moonshot AI | 1282 | 16,368 | Modified MIT |
| 60 | Kimi K2.5 (Instant) Moonshot AI | 1254 | 4,111 | Modified MIT |
Alternatives · Top open-weight models
| # | Model | Score | Votes | License |
|---|---|---|---|---|
| 24 |  GLM-5.3-Flash Z.ai | 1294 | 16,254 | MIT |
| 34 | Kimi K2.6 Moonshot AI | 1282 | 16,368 | Modified MIT |
| 38 |  DeepSeek-V4.1-Flash-20260910 (Max) DeepSeek | 1278 | 4,100 | MIT |
| 39 | Gemma 4 31B Google | 1276 | 40,219 | Apache 2.0 |
| 44 | Qwen3.8-27B Alibaba | 1269 | 7,614 | Apache 2.0 |
How to read these scores
- If two scores differ by less than their ± intervals, the models are statistically close.
- Each category board is computed independently; do not compare scores across categories.
- Models with few votes move more; ranks of new entries may change.
FAQ
Where does Kimi K2.5 (Thinking) rank among vision models?
Overall rank 46 (1269); OCR rank 47 (1276); Diagrams rank 50 (1282); Homework rank 38 (1297); Captioning rank 11 (1257); Entity recognition rank 7 (1285); Creative writing rank 40 (1269); Humor rank 24 (1284); English rank 44 (1274); Chinese rank 41 (1324) (LMArena, data as of Oct 26, 2026).
How good is Kimi K2.5 (Thinking) at OCR?
Kimi K2.5 (Thinking) ranks 47 of 123 on LMArena’s OCR board with 1276 (19,489 votes).
Is Kimi K2.5 (Thinking) better with Chinese or English prompts?
Chinese board: #41 of 128 (1324); English board: #44 of 167 (1274). The boards are computed separately, so compare ranks rather than scores.
Is Kimi K2.5 (Thinking) open-weight?
LMArena lists its license as Modified MIT, not proprietary; check the license text for commercial use.
How far is Kimi K2.5 (Thinking) from #1?
#1 overall is Claude Fable 5 (High) (1325); Kimi K2.5 (Thinking) has 1269, 56 points behind.
Sources
LMArena data published Oct 19, 2026.
Synced here Oct 26, 2026.
Price, context length and release date have no reliable public data here and are not shown; all figures come from snapshots, blanks are –.