# Qwen3.5 397B vs Qwen3.6-35B-A3B

> Side-by-side benchmark comparison for Qwen3.5 397B and Qwen3.6-35B-A3B across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/qwen3-5-397b-vs-qwen3-6-35b-a3b
- Last updated: October 14, 2026

- Shared sourced benchmarks: 40
- 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 55.63 to 43.54. Conditional score ranges do not establish rank confidence.
- The clearest category separation is in mathematics, where the averages are 81.7 for Qwen3.5 397B and 58.4 for Qwen3.6-35B-A3B.
- The biggest single benchmark swing is MCP Atlas in Agentic, with scores of 46.1% and 62.8%.
- Qwen3.6-35B-A3B also has the larger context window at 262K.

## Model Snapshot

| Property | Qwen3.5 397B | Qwen3.6-35B-A3B |
|----------|----------|----------|
| Creator | Alibaba | Alibaba |
| Type | Open Weight | Open Weight |
| Reasoning | Non-Reasoning | Reasoning |
| Context | 128K | 262K |
| Overall Score | 55.63 | 43.54 |
| Benchmarks Covered | 62 | 42 |
| Pricing (input/output) | $0.60 / $3.60 | N/A |

## Category Breakdown

### Agentic

- Winner: Directional only
- Qwen3.5 397B public-lane score: 44.3 (Estimated · #55/124)
- Qwen3.6-35B-A3B public-lane score: 29.2 (Estimated · #90/124)

| Benchmark | Qwen3.5 397B | Qwen3.6-35B-A3B | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 52.5% | 51.5% | Qwen3.5 397B |
| BrowseComp | 69% | Coming soon | Coming soon |
| OSWorld-Verified | 62.2% | Coming soon | Coming soon |
| Claw-Eval | 56.8% | 68.7% | Qwen3.6-35B-A3B |
| QwenClawBench | 51.8% | 52.6% | Qwen3.6-35B-A3B |
| τ³-bench results | 68.4% | 67.2% | Qwen3.5 397B |
| VITA-Bench | 43.7% | 35.6% | Qwen3.5 397B |
| DeepPlanning | 37.6% | 25.9% | Qwen3.5 397B |
| Toolathlon | 36.3% | 26.9% | Qwen3.5 397B |
| MCP Atlas | 46.1% | 62.8% | Qwen3.6-35B-A3B |
| MCP-Tasks | 74.2% | Coming soon | Coming soon |
| WideResearch | 74.0% | 60.1% | Qwen3.5 397B |
| Gert Labs | 46.76% | 42.65% | Qwen3.5 397B |
| ResearchClawBench | 14.2% | Coming soon | Coming soon |
| QwenWebBench | Coming soon | 1397 | Coming soon |

### Coding

- Winner: Not comparable
- Qwen3.5 397B public-lane score: Coming soon
- Qwen3.6-35B-A3B public-lane score: 31.8 (Estimated · #89/149)

| Benchmark | Qwen3.5 397B | Qwen3.6-35B-A3B | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 76.2% | 73.4% | Qwen3.5 397B |
| LiveCodeBench v6 | 83.6% | 80.4% | Qwen3.5 397B |
| SWE-bench Pro | 50.9% | 49.5% | Qwen3.5 397B |
| SWE Multilingual | 69.3% | 67.2% | Qwen3.5 397B |
| NL2Repo | 32.2% | 29.4% | Qwen3.5 397B |
| Terminal-Bench 2.0 | 52.5% | 51.5% | Qwen3.5 397B |

### Reasoning

- Winner: Not comparable
- Qwen3.5 397B public-lane score: 74.9 (Unranked · 2 rankable rows)
- Qwen3.6-35B-A3B public-lane score: 75.1 (Unranked · 2 rankable rows)

| Benchmark | Qwen3.5 397B | Qwen3.6-35B-A3B | Winner |
|-----------|-----------|-----------|--------|
| LongBench v2 | 63.2% | Coming soon | Coming soon |
| AI-Needle | 68.7% | Coming soon | Coming soon |

### Multimodal & Grounded

- Winner: Qwen3.5 397B
- Qwen3.5 397B public-lane score: 73.6 (#29/54)
- Qwen3.6-35B-A3B public-lane score: 66.4 (#40/54)

| Benchmark | Qwen3.5 397B | Qwen3.6-35B-A3B | Winner |
|-----------|-----------|-----------|--------|
| MMMU | 85.0% | 81.7% | Qwen3.5 397B |
| MMMU-Pro | 79% | 75.3% | Qwen3.5 397B |
| RealWorldQA | 83.9% | 85.3% | Qwen3.6-35B-A3B |
| OmniDocBench 1.5 | 90.8% | 89.9% | Qwen3.5 397B |
| Video-MME (with subtitle) | 87.5% | 86.6% | Qwen3.5 397B |
| Video-MME (w/o subtitle) | 83.7% | 82.5% | Qwen3.5 397B |
| MathVision | 88.6% | Coming soon | Coming soon |
| We-Math | 87.9% | Coming soon | Coming soon |
| DynaMath | 86.3% | Coming soon | Coming soon |
| SimpleVQA | 67.1% | 58.9% | Qwen3.5 397B |
| CharXiv | 80.8% | 78% | Qwen3.5 397B |
| MMLongBench-Doc | 61.5% | Coming soon | Coming soon |
| CC-OCR | 82.0% | 81.9% | Qwen3.5 397B |
| AI2D_TEST | 93.9% | 92.7% | Qwen3.5 397B |
| CountBench | 97.2% | Coming soon | Coming soon |
| RefCOCO (avg) | 92.3% | 92.0% | Qwen3.5 397B |
| ODINW13 | 47.0% | 50.8% | Qwen3.6-35B-A3B |
| ERQA | 67.5% | Coming soon | Coming soon |
| VideoMMMU | 84.7% | 83.7% | Qwen3.5 397B |
| MLVU (M-Avg) | 86.7% | 86.2% | Qwen3.5 397B |
| ScreenSpot Pro | 65.6% | Coming soon | Coming soon |
| V* | 95.8% | Coming soon | Coming soon |

### Knowledge

- Winner: Directional only
- Qwen3.5 397B public-lane score: 54.1 (Estimated · #64/181)
- Qwen3.6-35B-A3B public-lane score: 42.9 (Estimated · #105/181)

| Benchmark | Qwen3.5 397B | Qwen3.6-35B-A3B | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 88.4% | 86% | Qwen3.5 397B |
| SuperGPQA | 70.4% | 64.7% | Qwen3.5 397B |
| MMLU-Pro | 87.8% | 85.2% | Qwen3.5 397B |
| MMLU-Redux | 94.9% | Coming soon | Coming soon |
| C-Eval | 93% | 90% | Qwen3.5 397B |
| HLE | 28.7% | 21.4% | Qwen3.5 397B |

### Multilingual

- Winner: Not comparable
- Qwen3.5 397B public-lane score: 93.8 (#5/17)
- Qwen3.6-35B-A3B public-lane score: Coming soon

| Benchmark | Qwen3.5 397B | Qwen3.6-35B-A3B | Winner |
|-----------|-----------|-----------|--------|
| MMLU-ProX | 84.7% | Coming soon | Coming soon |
| NOVA-63 | 59.1% | Coming soon | Coming soon |
| INCLUDE | 85.6% | Coming soon | Coming soon |
| PolyMath | 73.3% | Coming soon | Coming soon |
| MAXIFE | 88.2% | Coming soon | Coming soon |

### Instruction Following

- Winner: Directional only
- Qwen3.5 397B public-lane score: 79.9 (#56/128)
- Qwen3.6-35B-A3B public-lane score: 78.3 (#58/128)

| Benchmark | Qwen3.5 397B | Qwen3.6-35B-A3B | Winner |
|-----------|-----------|-----------|--------|
| IFEval | 92.6% | Coming soon | Coming soon |
| IFBench | 76.5% | Coming soon | Coming soon |

### Mathematics

- Winner: Not comparable
- Qwen3.5 397B public-lane score: 81.7 (Unranked · 5 rankable rows)
- Qwen3.6-35B-A3B public-lane score: 58.4 (#10/12)

| Benchmark | Qwen3.5 397B | Qwen3.6-35B-A3B | Winner |
|-----------|-----------|-----------|--------|
| AIME26 | 93.3% | 92.7% | Qwen3.5 397B |
| HMMT Feb 2025 | 94.8% | 90.7% | Qwen3.5 397B |
| HMMT Nov 2025 | 92.7% | 89.1% | Qwen3.5 397B |
| HMMT Feb 2026 | 87.9% | 83.6% | Qwen3.5 397B |
| MMAnswerBench | 80.9% | 78.9% | Qwen3.5 397B |
| FrontierMath v2 (Tiers 1-3) | Coming soon | 20.351% | Coming soon |

## FAQ

### Which is better overall, Qwen3.5 397B or Qwen3.6-35B-A3B?

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

### Where is the biggest gap between Qwen3.5 397B and Qwen3.6-35B-A3B?

The widest category gap is in mathematics, where the averages are 81.7 for Qwen3.5 397B and 58.4 for Qwen3.6-35B-A3B.

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

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

### How many benchmarks does Qwen3.6-35B-A3B cover on BenchLM?

Qwen3.6-35B-A3B currently has 42 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Qwen3.5 397B vs Qwen3.5 397B (Reasoning)](/compare/qwen3-5-397b-vs-qwen3-5-397b-reasoning)
- [Qwen3.6-35B-A3B vs Qwen3.5 397B (Reasoning)](/compare/qwen3-5-397b-reasoning-vs-qwen3-6-35b-a3b)
- [Qwen3.5 397B vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-qwen3-5-397b)
- [Qwen3.6-35B-A3B vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-qwen3-6-35b-a3b)

## Explore More

- [Qwen3.5 397B profile](/models/qwen3-5-397b)
- [Qwen3.6-35B-A3B profile](/models/qwen3-6-35b-a3b)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)