Skip to main content

Model comparison

Qwen3 235B 2507 vs Qwen3.5 397B

Data verified

Head-to-head evidence from 4 shared benchmark results across 2 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

56.02/100
Margin
1.0pts
winning →
57.01/100
1 category wins1 category wins

Public leaderboard positions: Qwen3 235B 2507 #78 (Estimated); Qwen3.5 397B #71 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Qwen3 235B 2507 and Qwen3.5 397B share 4 comparable benchmark results. 2 of 8 categories are comparable. 0 results are unique to Qwen3 235B 2507; 51 to Qwen3.5 397B.

Updated July 22, 2026
Shared results
4
Qwen3 235B 2507 only
0
Qwen3.5 397B only
51
Comparable categories
2 / 8

Pick Qwen3.5 397B if you want the stronger benchmark profile. Qwen3 235B 2507 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 4 shared benchmark results across 2 evidence categories; 2 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 56.02. 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 multilingual, where it averages 84.7 against 79.4. The single biggest benchmark swing on the page is GPQA, 77.5% to 88.4%. Qwen3 235B 2507 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 235B 2507. That is roughly Infinityx on output cost alone.

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 scores and score margins for Qwen3 235B 2507 and Qwen3.5 397B
CategoryQwen3 235B 2507ΔQwen3.5 397B
KnowledgeQwen3 235B 250778.9Margin 22.3Qwen3.5 397B56.6
MultilingualQwen3 235B 250779.4Margin 5.3Qwen3.5 397B84.7
AgenticQwen3 235B 2507Not measuredMarginNo overlapQwen3.5 397B56.5
CodingQwen3 235B 2507Not measuredMarginNo overlapQwen3.5 397B66.5
ReasoningQwen3 235B 2507Not measuredMarginNo overlapQwen3.5 397B63.2
MathQwen3 235B 2507Not measuredMarginNo overlapQwen3.5 397B90.6
MultimodalQwen3 235B 2507Not measuredMarginNo overlapQwen3.5 397B79.6
Inst. FollowingQwen3 235B 2507Not measuredMarginNo overlapQwen3.5 397B92.6

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · Qwen3 235B 2507B · Qwen3.5 397B
  1. GPQA

    Knowledge
    Source ↗
    A 77.5%B 88.4%
    Winner: Qwen3.5 397BΔ 10.9
    GPQA: Qwen3 235B 2507 scored 77.5%; Qwen3.5 397B scored 88.4%. Qwen3.5 397B wins this benchmark.
  2. SuperGPQA

    Knowledge
    Source ↗
    A 62.6%B 70.4%
    Winner: Qwen3.5 397BΔ 7.8
    SuperGPQA: Qwen3 235B 2507 scored 62.6%; Qwen3.5 397B scored 70.4%. Qwen3.5 397B wins this benchmark.
  3. MMLU-ProX

    Multilingual
    Source ↗
    A 79.4%B 84.7%
    Winner: Qwen3.5 397BΔ 5.3
    MMLU-ProX: Qwen3 235B 2507 scored 79.4%; Qwen3.5 397B scored 84.7%. Qwen3.5 397B wins this benchmark.
  4. MMLU-Pro

    Knowledge
    Source ↗
    A 83%B 87.8%
    Winner: Qwen3.5 397BΔ 4.8
    MMLU-Pro: Qwen3 235B 2507 scored 83%; Qwen3.5 397B scored 87.8%. Qwen3.5 397B wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricQwen3 235B 2507Qwen3.5 397BComparison
Input / output priceUSD per 1M tokensQwen3 235B 2507$0 input / $0 outputQwen3.5 397B$0.6 input / $3.6 outputQwen3 235B 2507 has the lower combined listed price.
Generation speedtokens per secondQwen3 235B 2507Not availableQwen3.5 397B96 tok/sA complete speed comparison is not available.
First-answer latencyseconds to first tokenQwen3 235B 2507Not availableQwen3.5 397B2.44 sA complete latency comparison is not available.
Context windowmaximum listed tokensQwen3 235B 2507128KQwen3.5 397B128KListed context windows are equal.

Benchmark Deep Dive

Agentic
BenchmarkQwen3 235B 2507Qwen3.5 397BResult
Terminal-Bench 2.0Source 52.5%Not comparable
BrowseCompSource 62%Not comparable
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
τ²-bench resultsSource 95.6%Not comparable
Gert LabsSource 46.76%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 962Not comparable
Coding
BenchmarkQwen3 235B 2507Qwen3.5 397BResult
SWE-bench VerifiedSource 76.2%Not comparable
LiveCodeBench v6Source 83.6%Not comparable
SWE-bench ProSource 50.9%Not comparable
AA-SciCodeSource 42.0%Not comparable
AA Coding IndexSource 48.2%Not comparable
Reasoning
BenchmarkQwen3 235B 2507Qwen3.5 397BResult
LongBench v2Source 63.2%Not comparable
AI-NeedleSource 68.7%Not comparable
AA-LCRSource 65.7%Not comparable
CritPtSource 1.7%Not comparable
KnowledgeQwen3 235B 2507 wins
BenchmarkQwen3 235B 2507Qwen3.5 397BResult
GPQASource 77.5%88.4%Qwen3.5 397B leads
SuperGPQASource 62.6%70.4%Qwen3.5 397B leads
MMLU-ProSource 83%87.8%Qwen3.5 397B leads
MMLU-ReduxSource 94.9%Not comparable
C-EvalSource 93%Not comparable
HLESource 28.7%Not comparable
Artificial Analysis Intelligence IndexSource 33.7%Not comparable
AA-GPQA DiamondSource 89.3%Not comparable
AA-HLESource 27.3%Not comparable
AA-Omniscience IndexSource -29.8%Not comparable
AA-Omniscience AccuracySource 31.4%Not comparable
AA-Omniscience Hallucination RateSource 89.1%Not comparable
Math
BenchmarkQwen3 235B 2507Qwen3.5 397BResult
AIME26Source 93.3%Not comparable
HMMT Feb 2025Source 94.8%Not comparable
HMMT Nov 2025Source 92.7%Not comparable
HMMT Feb 2026Source 87.9%Not comparable
MMAnswerBenchSource 80.9%Not comparable
MultilingualQwen3.5 397B wins
BenchmarkQwen3 235B 2507Qwen3.5 397BResult
MMLU-ProXSource 79.4%84.7%Qwen3.5 397B leads
NOVA-63Source 59.1%Not comparable
Multimodal
BenchmarkQwen3 235B 2507Qwen3.5 397BResult
MMMU-ProSource 79%Not comparable
MathVisionSource 88.6%Not comparable
CharXivSource 80.8%Not comparable
VideoMMMUSource 84.7%Not comparable
ScreenSpot ProSource 65.6%Not comparable
V*Source 95.8%Not comparable
AA-MMMU-ProSource 77.3%Not comparable
Inst. Following
BenchmarkQwen3 235B 2507Qwen3.5 397BResult
IFEvalSource 92.6%Not comparable
AA-IFBenchSource 78.8%Not comparable
Frequently Asked Questions (3)

Which is better, Qwen3 235B 2507 or Qwen3.5 397B?

Qwen3.5 397B is ahead on BenchLM's BenchAlign leaderboard, 57.01 to 56.02. The biggest single separator in this matchup is GPQA, where the scores are 77.5% and 88.4%.

Which is better for knowledge tasks, Qwen3 235B 2507 or Qwen3.5 397B?

Qwen3 235B 2507 has the edge for knowledge tasks in this comparison, averaging 78.9 versus 56.6. Inside this category, GPQA is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, Qwen3 235B 2507 or Qwen3.5 397B?

Qwen3.5 397B has the edge for multilingual tasks in this comparison, averaging 84.7 versus 79.4. Inside this category, MMLU-ProX is the benchmark that creates the most daylight between them.

Related Comparisons

Last updated: July 22, 2026

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.