Model comparison
Qwen3.5-35B-A3B vs Qwen3.5 397B
Head-to-head evidence from 24 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Qwen3.5-35B-A3B #72 (Supported); Qwen3.5 397B #71 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Qwen3.5-35B-A3B and Qwen3.5 397B share 24 comparable benchmark results. 6 of 8 categories are comparable. 4 results are unique to Qwen3.5-35B-A3B; 31 to Qwen3.5 397B.
Updated July 23, 2026- Shared results
- 24
- Qwen3.5-35B-A3B only
- 4
- Qwen3.5 397B only
- 31
- Comparable categories
- 6 / 8
Pick Qwen3.5 397B if you want the stronger benchmark profile. Qwen3.5-35B-A3B only becomes the better choice if knowledge is the priority or you want the cheaper token bill.
Confidence note. This is a partial-evidence comparison with 24 shared benchmark results across 7 evidence categories; 6 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
Qwen3.5 397B has the cleaner BenchAlign overall profile here, landing at 57.01 versus 56.97. It is a real lead, but still close enough that category-level strengths matter more than the headline number.
Qwen3.5 397B's sharpest advantage is in coding, where it averages 66.5 against 60.6. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 40.5% to 52.5%. Qwen3.5-35B-A3B does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.
Qwen3.5 397B is also the more expensive model on tokens at $0.60 input / $3.60 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Qwen3.5-35B-A3B. That is roughly Infinityx on output cost alone. Qwen3.5-35B-A3B is the reasoning model in the pair, while Qwen3.5 397B is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Qwen3.5-35B-A3B gives you the larger context window at 262K, compared with 128K for Qwen3.5 397B.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | Qwen3.5-35B-A3B | Δ | Qwen3.5 397B |
|---|---|---|---|
| Knowledge | Qwen3.5-35B-A3B81.6 | Margin← 25.0 | Qwen3.5 397B56.6 |
| Coding | Qwen3.5-35B-A3B60.6 | Margin→ 5.9 | Qwen3.5 397B66.5 |
| Agentic | Qwen3.5-35B-A3B51.0 | Margin→ 5.5 | Qwen3.5 397B56.5 |
| Reasoning | Qwen3.5-35B-A3B59.0 | Margin→ 4.2 | Qwen3.5 397B63.2 |
| Multilingual | Qwen3.5-35B-A3B81.0 | Margin→ 3.7 | Qwen3.5 397B84.7 |
| Inst. Following | Qwen3.5-35B-A3B91.9 | Margin→ 0.7 | Qwen3.5 397B92.6 |
| Math | Qwen3.5-35B-A3BNot measured | MarginNo overlap | Qwen3.5 397B90.6 |
| Multimodal | Qwen3.5-35B-A3BNot measured | MarginNo overlap | Qwen3.5 397B79.6 |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
Terminal-Bench 2.0
AgenticA 40.5%B 52.5%Winner: Qwen3.5 397BΔ 12Terminal-Bench 2.0: Qwen3.5-35B-A3B scored 40.5%; Qwen3.5 397B scored 52.5%. Qwen3.5 397B wins this benchmark. - Source ↗
SuperGPQA
KnowledgeA 63.4%B 70.4%Winner: Qwen3.5 397BΔ 7SuperGPQA: Qwen3.5-35B-A3B scored 63.4%; Qwen3.5 397B scored 70.4%. Qwen3.5 397B wins this benchmark. - Source ↗
SWE-bench Verified
CodingA 69.2%B 76.2%Winner: Qwen3.5 397BΔ 7SWE-bench Verified: Qwen3.5-35B-A3B scored 69.2%; Qwen3.5 397B scored 76.2%. Qwen3.5 397B wins this benchmark. - Source ↗
LongBench v2
ReasoningA 59%B 63.2%Winner: Qwen3.5 397BΔ 4.2LongBench v2: Qwen3.5-35B-A3B scored 59%; Qwen3.5 397B scored 63.2%. Qwen3.5 397B wins this benchmark. - Source ↗
GPQA
KnowledgeA 84.2%B 88.4%Winner: Qwen3.5 397BΔ 4.2GPQA: Qwen3.5-35B-A3B scored 84.2%; Qwen3.5 397B scored 88.4%. Qwen3.5 397B wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Qwen3.5-35B-A3B | Qwen3.5 397B | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Qwen3.5-35B-A3B$0 input / $0 output | Qwen3.5 397B$0.6 input / $3.6 output | Qwen3.5-35B-A3B has the lower combined listed price. |
| Generation speedtokens per second | Qwen3.5-35B-A3BNot available | Qwen3.5 397B96 tok/s | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Qwen3.5-35B-A3BNot available | Qwen3.5 397B2.44 s | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Qwen3.5-35B-A3B262K | Qwen3.5 397B128K | Qwen3.5-35B-A3B lists the larger context window. |
Benchmark Deep Dive
AgenticQwen3.5 397B wins19 benchmarks
| Benchmark | Qwen3.5-35B-A3B | Qwen3.5 397B | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 40.5% | 52.5% | Qwen3.5 397B leads |
| BrowseCompSource | 61% | 62% | Qwen3.5 397B leads |
| OSWorld-VerifiedSource | 54.5% | — | Not comparable |
| τ²-bench resultsSource | 89.2% | 95.6% | Qwen3.5 397B leads |
| Gert LabsSource | 28.96% | 46.76% | Qwen3.5 397B leads |
| Claw-EvalSource | — | 56.8% | Not comparable |
| QwenClawBenchSource | — | 51.8% | Not comparable |
| τ³-bench resultsSource | — | 68.4% | Not comparable |
| VITA-BenchSource | — | 43.7% | Not comparable |
| DeepPlanningSource | — | 37.6% | Not comparable |
| ToolathlonSource | — | 36.3% | Not comparable |
| MCP AtlasSource | — | 46.1% | Not comparable |
| MCP-TasksSource | — | 74.2% | Not comparable |
| WideResearchSource | — | 74.0% | Not comparable |
| ResearchClawBenchSource | — | 14.2% | Not comparable |
| AA Agentic IndexSource | — | 19.9% | Not comparable |
| APEX-Agents-AASource | — | 15.3% | Not comparable |
| GDPval-AASource | — | 23.1% | Not comparable |
| GDPval-AASource | — | 962 | Not comparable |
CodingQwen3.5 397B wins6 benchmarks
| Benchmark | Qwen3.5-35B-A3B | Qwen3.5 397B | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 69.2% | 76.2% | Qwen3.5 397B leads |
| SWE-RebenchSource | 53.7% | — | Not comparable |
| AA-SciCodeSource | 37.7% | 42.0% | Qwen3.5 397B leads |
| LiveCodeBench v6Source | — | 83.6% | Not comparable |
| SWE-bench ProSource | — | 50.9% | Not comparable |
| AA Coding IndexSource | — | 48.2% | Not comparable |
ReasoningQwen3.5 397B wins4 benchmarks
KnowledgeQwen3.5-35B-A3B wins12 benchmarks
| Benchmark | Qwen3.5-35B-A3B | Qwen3.5 397B | Result |
|---|---|---|---|
| MMLU-ProSource | 85.3% | 87.8% | Qwen3.5 397B leads |
| SuperGPQASource | 63.4% | 70.4% | Qwen3.5 397B leads |
| GPQASource | 84.2% | 88.4% | Qwen3.5 397B leads |
| Artificial Analysis Intelligence IndexSource | 29.3% | 33.7% | Qwen3.5 397B leads |
| AA-GPQA DiamondSource | 84.5% | 89.3% | Qwen3.5 397B leads |
| AA-HLESource | 19.7% | 27.3% | Qwen3.5 397B leads |
| AA-Omniscience IndexSource | -46.4% | -29.8% | Qwen3.5 397B leads |
| AA-Omniscience AccuracySource | 20.5% | 31.4% | Qwen3.5 397B leads |
| AA-Omniscience Hallucination RateSource | 84.0% | 89.1% | Qwen3.5-35B-A3B leads |
| MMLU-ReduxSource | — | 94.9% | Not comparable |
| C-EvalSource | — | 93% | Not comparable |
| HLESource | — | 28.7% | Not comparable |
Math5 benchmarks
MultilingualQwen3.5 397B wins2 benchmarks
Multimodal9 benchmarks
| Benchmark | Qwen3.5-35B-A3B | Qwen3.5 397B | Result |
|---|---|---|---|
| MMMUSource | 81.4% | — | Not comparable |
| MMVUSource | 72.3% | — | Not comparable |
| MathVisionSource | 83.9% | 88.6% | Qwen3.5 397B leads |
| V*Source | 92.7% | 95.8% | Qwen3.5 397B leads |
| AA-MMMU-ProSource | 72.7% | 77.3% | Qwen3.5 397B leads |
| MMMU-ProSource | — | 79% | Not comparable |
| CharXivSource | — | 80.8% | Not comparable |
| VideoMMMUSource | — | 84.7% | Not comparable |
| ScreenSpot ProSource | — | 65.6% | Not comparable |
Frequently Asked Questions (7)
Which is better, Qwen3.5-35B-A3B or Qwen3.5 397B?
Qwen3.5 397B is ahead on BenchLM's BenchAlign leaderboard, 57.01 to 56.97. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 40.5% and 52.5%.
Which is better for knowledge tasks, Qwen3.5-35B-A3B or Qwen3.5 397B?
Qwen3.5-35B-A3B has the edge for knowledge tasks in this comparison, averaging 81.6 versus 56.6. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.
Which is better for coding, Qwen3.5-35B-A3B or Qwen3.5 397B?
Qwen3.5 397B has the edge for coding in this comparison, averaging 66.5 versus 60.6. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.
Which is better for reasoning, Qwen3.5-35B-A3B or Qwen3.5 397B?
Qwen3.5 397B has the edge for reasoning in this comparison, averaging 63.2 versus 59. Inside this category, LongBench v2 is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, Qwen3.5-35B-A3B or Qwen3.5 397B?
Qwen3.5 397B has the edge for agentic tasks in this comparison, averaging 56.5 versus 51. Inside this category, Gert Labs is the benchmark that creates the most daylight between them.
Which is better for instruction following, Qwen3.5-35B-A3B or Qwen3.5 397B?
Qwen3.5 397B has the edge for instruction following in this comparison, averaging 92.6 versus 91.9. Inside this category, AA-IFBench is the benchmark that creates the most daylight between them.
Which is better for multilingual tasks, Qwen3.5-35B-A3B or Qwen3.5 397B?
Qwen3.5 397B has the edge for multilingual tasks in this comparison, averaging 84.7 versus 81. Inside this category, MMLU-ProX is the benchmark that creates the most daylight between them.
Related Comparisons
Explore More
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.