# Kimi K2.5 vs Qwen3.5 397B

> Side-by-side benchmark comparison for Kimi K2.5 and Qwen3.5 397B across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/kimi-k2-5-vs-qwen3-5-397b
- Last updated: October 17, 2026

- Shared sourced benchmarks: 55
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

Qwen3.5 397B leads by point estimate. Conditional score ranges do not establish rank confidence. Choose using the category evidence and your own workload tests.

## Summary

- Qwen3.5 397B has the higher overall point estimate 55.34 to 53.05. Conditional score ranges do not establish rank confidence.
- The clearest category separation is in multilingual, where the averages are 87.3 for Kimi K2.5 and 93.8 for Qwen3.5 397B.
- The biggest single benchmark swing is PolyMath in Multilingual, with scores of 43.1% and 73.3%.
- Kimi K2.5 is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- Kimi K2.5 also has the larger context window at 256K.

## Model Snapshot

| Property | Kimi K2.5 | Qwen3.5 397B |
|----------|----------|----------|
| Creator | Moonshot AI | Alibaba |
| Type | Open Weight | Open Weight |
| Reasoning | Non-Reasoning | Non-Reasoning |
| Context | 256K | 128K |
| Overall Score | 53.05 | 55.34 |
| Benchmarks Covered | 68 | 62 |
| Pricing (input/output) | $0.60 / $3.00 | $0.60 / $3.60 |

## Category Breakdown

### Agentic

- Winner: Directional only
- Kimi K2.5 public-lane score: 40.7 (Estimated · #66/128)
- Qwen3.5 397B public-lane score: 43.8 (Estimated · #61/128)

| Benchmark | Kimi K2.5 | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 50.8% | 52.5% | Qwen3.5 397B |
| BrowseComp | 60.6% | 69% | Qwen3.5 397B |
| OSWorld-Verified | 63.3% | 62.2% | Kimi K2.5 |
| Claw-Eval | 52.3% | 56.8% | Qwen3.5 397B |
| QwenClawBench | 54.3% | 51.8% | Kimi K2.5 |
| τ³-bench results | 65.7% | 68.4% | Qwen3.5 397B |
| DeepSearchQA | 77.1% | Coming soon | Coming soon |
| DeepPlanning | 14.4% | 37.6% | Qwen3.5 397B |
| Toolathlon | 27.8% | 36.3% | Qwen3.5 397B |
| MCP Atlas | 29.5% | 46.1% | Qwen3.5 397B |
| MCP-Tasks | 59.1% | 74.2% | Qwen3.5 397B |
| WideResearch | 72.7% | 74.0% | Qwen3.5 397B |
| BFCL v4 | 68.3% | Coming soon | Coming soon |
| Gert Labs | 45.88% | 46.76% | Qwen3.5 397B |
| ResearchClawBench | 14.0% | 14.2% | Qwen3.5 397B |
| JobBench | 8.7% | Coming soon | Coming soon |
| VITA-Bench | Coming soon | 43.7% | Coming soon |

### Coding

- Winner: Not comparable
- Kimi K2.5 public-lane score: 38.5 (Estimated · #68/150)
- Qwen3.5 397B public-lane score: Coming soon

| Benchmark | Kimi K2.5 | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 76.8% | 76.2% | Kimi K2.5 |
| SWE-bench Verified* | 70.8% | Coming soon | Coming soon |
| LiveCodeBench v6 | 85.0% | 83.6% | Kimi K2.5 |
| SWE-bench Pro | 50.7% | 50.9% | Qwen3.5 397B |
| SWE Multilingual | 73% | 69.3% | Kimi K2.5 |
| SWE-Rebench | 58.5% | Coming soon | Coming soon |
| React Native Evals | 77.2% | Coming soon | Coming soon |
| NL2Repo | 32% | 32.2% | Qwen3.5 397B |
| SciCode | 48.7% | Coming soon | Coming soon |
| Terminal-Bench 2.0 | 50.8% | 52.5% | Qwen3.5 397B |

### Reasoning

- Winner: Not comparable
- Kimi K2.5 public-lane score: 72.2 (Unranked · 3 rankable rows)
- Qwen3.5 397B public-lane score: 74.9 (Unranked · 2 rankable rows)

| Benchmark | Kimi K2.5 | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| LongBench v2 | 61% | 63.2% | Qwen3.5 397B |
| AI-Needle | Coming soon | 68.7% | Coming soon |

### Multimodal & Grounded

- Winner: Qwen3.5 397B
- Kimi K2.5 public-lane score: 70.4 (#32/54)
- Qwen3.5 397B public-lane score: 73.6 (#29/54)

| Benchmark | Kimi K2.5 | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| MMMU | 84.3% | 85.0% | Qwen3.5 397B |
| MMMU-Pro | 78.5% | 79% | Qwen3.5 397B |
| RealWorldQA | 81.0% | 83.9% | Qwen3.5 397B |
| OmniDocBench 1.5 | 88.8% | 90.8% | Qwen3.5 397B |
| Video-MME | 87.4% | Coming soon | Coming soon |
| Video-MME (with subtitle) | 87.4% | 87.5% | Qwen3.5 397B |
| Video-MME (w/o subtitle) | 83.2% | 83.7% | Qwen3.5 397B |
| MathVision | 84.2% | 88.6% | Qwen3.5 397B |
| We-Math | 84.7% | 87.9% | Qwen3.5 397B |
| DynaMath | 84.4% | 86.3% | Qwen3.5 397B |
| SimpleVQA | 71.2% | 67.1% | Kimi K2.5 |
| MMLongBench-Doc | 58.5% | 61.5% | Qwen3.5 397B |
| CC-OCR | 79.7% | 82.0% | Qwen3.5 397B |
| AI2D_TEST | 90.8% | 93.9% | Qwen3.5 397B |
| CountBench | 94.1% | 97.2% | Qwen3.5 397B |
| RefCOCO (avg) | 87.8% | 92.3% | Qwen3.5 397B |
| MMVU | 80.4% | Coming soon | Coming soon |
| VideoMMMU | 86.6% | 84.7% | Kimi K2.5 |
| MLVU (M-Avg) | 85.0% | 86.7% | Qwen3.5 397B |
| CharXiv | 77.5% | 80.8% | Qwen3.5 397B |
| ODINW13 | Coming soon | 47.0% | Coming soon |
| ERQA | Coming soon | 67.5% | Coming soon |
| ScreenSpot Pro | Coming soon | 65.6% | Coming soon |
| V* | Coming soon | 95.8% | Coming soon |

### Knowledge

- Winner: Directional only
- Kimi K2.5 public-lane score: 51.4 (Estimated · #75/184)
- Qwen3.5 397B public-lane score: 53.1 (Estimated · #68/184)

| Benchmark | Kimi K2.5 | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 87.6% | 88.4% | Qwen3.5 397B |
| GPQA-D | 87.6% | Coming soon | Coming soon |
| SuperGPQA | 69.2% | 70.4% | Qwen3.5 397B |
| MMLU-Pro | 87.1% | 87.8% | Qwen3.5 397B |
| MMLU-Pro (Arcee) | 87.1% | Coming soon | Coming soon |
| MMLU-Redux | 94.5% | 94.9% | Qwen3.5 397B |
| C-Eval | 94% | 93% | Kimi K2.5 |
| HLE | 30.1% | 28.7% | Kimi K2.5 |

### Multilingual

- Winner: Qwen3.5 397B
- Kimi K2.5 public-lane score: 87.3 (#8/17)
- Qwen3.5 397B public-lane score: 93.8 (#5/17)

| Benchmark | Kimi K2.5 | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| MMLU-ProX | 82.3% | 84.7% | Qwen3.5 397B |
| NOVA-63 | 56.0% | 59.1% | Qwen3.5 397B |
| INCLUDE | 83.3% | 85.6% | Qwen3.5 397B |
| PolyMath | 43.1% | 73.3% | Qwen3.5 397B |
| MAXIFE | 72.8% | 88.2% | Qwen3.5 397B |

### Instruction Following

- Winner: Directional only
- Kimi K2.5 public-lane score: 85.8 (#48/148)
- Qwen3.5 397B public-lane score: 79.9 (#59/148)

| Benchmark | Kimi K2.5 | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| IFEval | 93.9% | 92.6% | Kimi K2.5 |
| IFBench | Coming soon | 76.5% | Coming soon |

### Mathematics

- Winner: Not comparable
- Kimi K2.5 public-lane score: 83.3 (Unranked · 7 rankable rows)
- Qwen3.5 397B public-lane score: 81.7 (Unranked · 5 rankable rows)

| Benchmark | Kimi K2.5 | Qwen3.5 397B | Winner |
|-----------|-----------|-----------|--------|
| AIME 2025 | 96.1% | Coming soon | Coming soon |
| AIME26 | 95.8% | 93.3% | Kimi K2.5 |
| AIME25 (Arcee) | 96.3% | Coming soon | Coming soon |
| HMMT Feb 2025 | 95.4% | 94.8% | Kimi K2.5 |
| HMMT Nov 2025 | 91.1% | 92.7% | Qwen3.5 397B |
| HMMT Feb 2026 | 87.1% | 87.9% | Qwen3.5 397B |
| MMAnswerBench | 81.8% | 80.9% | Kimi K2.5 |

## FAQ

### Which is better overall, Kimi K2.5 or Qwen3.5 397B?

Qwen3.5 397B has the higher overall point estimate. Conditional score ranges do not establish rank confidence.

### Where is the biggest gap between Kimi K2.5 and Qwen3.5 397B?

The widest category gap is in multilingual, where the averages are 87.3 for Kimi K2.5 and 93.8 for Qwen3.5 397B.

### How many benchmarks does Kimi K2.5 cover on BenchLM?

Kimi K2.5 currently has 68 sourced benchmark scores on BenchLM.

### How many benchmarks does Qwen3.5 397B cover on BenchLM?

Qwen3.5 397B currently has 62 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Kimi K2.5 vs Kimi K2.5 (Reasoning)](/compare/kimi-k2-5-vs-kimi-k2-5-reasoning)
- [Qwen3.5 397B vs Kimi K2.5 (Reasoning)](/compare/kimi-k2-5-reasoning-vs-qwen3-5-397b)
- [Kimi K2.5 vs Qwen3.5 397B (Reasoning)](/compare/kimi-k2-5-vs-qwen3-5-397b-reasoning)
- [Qwen3.5 397B vs Qwen3.5 397B (Reasoning)](/compare/qwen3-5-397b-vs-qwen3-5-397b-reasoning)

## Explore More

- [Kimi K2.5 profile](/models/kimi-k2-5)
- [Qwen3.5 397B profile](/models/qwen3-5-397b)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)