DeepSeek V3.2 vs Kimi K2.5
Head to head on 25 shared benchmarks, with price, context window, and release dates.
Benchmark scores
45evals · sort by any column, ⤢ to expand fullscreen
| Model | ||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Kimi K2.5 Moonshot AI | 68.3% | 35.0% | 1579elo | 74.9 | 27.9% | 87.9% | 30.7% | 70.2% | 78.5% | 72.5% | 77.9% | 61.4% | 57.4% | 77.7% | 84.9% | 76.0% | 62.0% | 49.0% | 67.3% | 70.8% | 74.2% | 34.8% | 69.4 | 17.5% | 95.9% | - | - | - | 96.1 | - | 35.8% | 95.4 | - | 10421 | - | 39.3% | 76.4% | 87.1 | - | 66.5% | 63.3% | - | 49.9% | 73 | 26.3% | 59.8% |
| DeepSeek V3.2 DeepSeek | 47.9% | 13.0% | 1515elo | 66.8 | 22.1% | 84.0% | 24.6% | 60.7% | 68.0% | 65.9% | 64.6% | 50.0% | 48.2% | 70.4% | 85.0% | 77.2% | 57.5% | 38.9% | 59.0% | 70.0% | 68.2% | 35.6% | 65 | 5.1% | 90.6% | 74.7% | 74.2% | 92.0% | - | 0.0% | - | - | 31.3% | - | 86.2% | - | - | - | 86.2% | - | - | 8.0% | - | - | - | 55.3% |
2 / 2 models
| Attribute | DeepSeek V3.2 DeepSeek | Kimi K2.5 Moonshot AI |
|---|---|---|
| Description | DeepSeek V3.2 is an AI model from DeepSeek, released with open weights. | Kimi K2.5 is an AI model from Kimi. |
| Family | Deepseek | Kimi |
| Params | - | - |
| Open weights | open | open |
| License | mit | Modified MIT |
| Released | 1 Dec 2025 | 27 Jan 2026 |
| Context | 163,840 | 262,144 |
| Input $/1M | $0.28 | $0.60 |
| Output $/1M | $0.42 | $2.75 |
| Cache read $/1M | $0.135 | $0.070 |
| Cache write $/1M | - | - |
| Speed | 0 tok/s | 0 tok/s |
| Latency | 0.00 s | 0.00 s |
| Modalities | text | text, image |
| Providers | - | - |
| Publisher | DeepSeek | Moonshot AI |
| Reported scores | Aider Polyglot Benchmark74.2% (Aider) MathArena57.47% GPQA Diamond84.0% (AA) Humanity's Last Exam (HLE)24.6% (AA) | GPQA Diamond87.9% (AA) Humanity's Last Exam (HLE)30.7% (AA) IFBench70.2% (AA) SciCode49.0% (AA) τ²-bench (Tau²-bench)95.9% (AA) |
FAQ
- Is DeepSeek V3.2 or Kimi K2.5 better?
- Across the 25 benchmarks both models report on Sophon, Kimi K2.5 leads on 22 of them. Which one is "better" depends on the benchmark - the table above shows every shared score.
- How do DeepSeek V3.2 and Kimi K2.5 compare on GPQA Diamond?
- DeepSeek V3.2 scores 84.0% and Kimi K2.5 scores 87.9% on GPQA Diamond.
- Which is cheaper, DeepSeek V3.2 or Kimi K2.5?
- DeepSeek V3.2 is cheaper at $0.42 per million output tokens against $2.75 for Kimi K2.5 - about 6.5x.
- When were DeepSeek V3.2 and Kimi K2.5 released?
- DeepSeek V3.2 was released 1 Dec 2025 by DeepSeek. Kimi K2.5 was released 27 Jan 2026 by Moonshot AI.
Other head-to-heads
56Claude Opus 4.6 vs Kimi K2.531 shared evalsGemini 3.1 Pro Preview vs Kimi K2.530 shared evalsGPT-5.2 vs Kimi K2.529 shared evalsDeepSeek V3.2 vs GPT-5.229 shared evalsKimi K2.5 vs Qwen3.6 Plus29 shared evalsClaude Sonnet 4.5 vs Kimi K2.528 shared evalsGPT-5.5 vs Kimi K2.528 shared evalsClaude Opus 4.8 vs Kimi K2.528 shared evalsGPT-5.4 vs Kimi K2.528 shared evalsClaude Opus 4.7 vs Kimi K2.528 shared evalsClaude Sonnet 4.5 vs DeepSeek V3.227 shared evalsClaude Opus 4.5 vs DeepSeek V3.227 shared evalsClaude Sonnet 4.6 vs Kimi K2.527 shared evalsDeepSeek V3.2 vs GPT-5 Mini26 shared evalsClaude Opus 4.5 vs Kimi K2.526 shared evalsClaude Fable 5 vs Kimi K2.526 shared evalsClaude Opus 4.6 vs DeepSeek V3.226 shared evalsGLM 5.1 vs Kimi K2.526 shared evalsDeepSeek V3.2 vs Grok 4.1 Fast26 shared evalsGPT-5 Mini vs Kimi K2.525 shared evalsGPT-5.4 Mini vs Kimi K2.525 shared evalsDeepSeek V3.2 vs Gemini 3.1 Pro Preview24 shared evalsDeepSeek V3.2 vs GPT-5.424 shared evalsGrok 4.1 Fast vs Kimi K2.524 shared evalsDeepSeek V3.2 vs GLM 5.124 shared evalsDeepSeek V3.2 vs GPT-5.523 shared evalsClaude 4 Sonnet vs Kimi K2.523 shared evalsClaude Sonnet 4.6 vs DeepSeek V3.223 shared evalsClaude Opus 4.7 vs DeepSeek V3.223 shared evalsDeepSeek V3.2 vs gpt-oss-120b23 shared evalsGPT-5 Nano vs Kimi K2.522 shared evalsDeepSeek V3.2 vs GPT-5 Nano22 shared evalsClaude Opus 4.8 vs DeepSeek V3.222 shared evalsDeepSeek V3.2 vs GLM 4.722 shared evalsDeepSeek V3.2 vs GPT-5.4 Mini22 shared evalsClaude 4 Sonnet vs DeepSeek V3.221 shared evalsGemini 2.5 Pro vs Kimi K2.521 shared evalsDeepSeek V3.2 vs Gemini 2.5 Pro21 shared evalsClaude Fable 5 vs DeepSeek V3.220 shared evalsGLM 4.7 vs Kimi K2.520 shared evalsDeepSeek V3.2 vs Qwen3.6 Plus20 shared evalsDeepSeek V3.2 vs GPT-519 shared evalsgpt-oss-120b vs Kimi K2.519 shared evalsGPT-5 vs Kimi K2.518 shared evalsDeepSeek V3.2 vs GPT-4o18 shared evalsDeepSeek V3.2 vs GPT-4.1 Mini15 shared evalsDeepSeek V3.2 vs GPT-4.115 shared evalsGPT-4o vs Kimi K2.515 shared evalsDeepSeek V3.2 vs GPT-4o-mini15 shared evalsDeepSeek V3.2 vs GPT-4.1 Nano15 shared evalsGPT-6 Astra vs Kimi K2.513 shared evalsGPT-4.1 Mini vs Kimi K2.512 shared evalsGPT-4.1 vs Kimi K2.512 shared evalsGPT-4o-mini vs Kimi K2.512 shared evalsGPT-4.1 Nano vs Kimi K2.512 shared evalsDeepSeek V3.2 vs GPT-6 Astra9 shared evals
Comparing something else? Build your own side-by-side.