Model comparison
GPT-5.4 vs Qwen3.6-35B-A3B
Head-to-head evidence from 28 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Verified leaderboard positions: GPT-5.4 #10; Qwen3.6-35B-A3B #31
BenchAlign evidence: GPT-5.4 supported; Qwen3.6-35B-A3B estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. GPT-5.4 and Qwen3.6-35B-A3B share 28 comparable benchmark results. 5 of 8 categories are comparable. 26 results are unique to GPT-5.4; 30 to Qwen3.6-35B-A3B.
Updated July 15, 2026- Shared results
- 28
- GPT-5.4 only
- 26
- Qwen3.6-35B-A3B only
- 30
- Comparable categories
- 5 / 8
Pick GPT-5.4 if you want the stronger benchmark profile. Qwen3.6-35B-A3B only becomes the better choice if mathematics is the priority.
Confidence note. This is a partial-evidence comparison with 28 shared benchmark results across 6 evidence categories; 5 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
GPT-5.4 is clearly ahead on the provisional aggregate, 86 to 59. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
GPT-5.4's sharpest advantage is in agentic, where it averages 77.2 against 51.5. The single biggest benchmark swing on the page is HLE, 52.1% to 21.4%. Qwen3.6-35B-A3B does hit back in mathematics, so the answer changes if that is the part of the workload you care about most.
GPT-5.4 gives you the larger context window at 1.05M, compared with 262K for Qwen3.6-35B-A3B.
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 | GPT-5.4 | Δ | Qwen3.6-35B-A3B |
|---|---|---|---|
| Math | GPT-5.442.5 | Margin→ 45.7 | Qwen3.6-35B-A3B88.2 |
| Agentic | GPT-5.477.2 | Margin← 25.7 | Qwen3.6-35B-A3B51.5 |
| Coding | GPT-5.457.7 | Margin→ 16.1 | Qwen3.6-35B-A3B73.8 |
| Knowledge | GPT-5.457.6 | Margin← 5.8 | Qwen3.6-35B-A3B51.8 |
| Multimodal | GPT-5.473.2 | Margin→ 3.1 | Qwen3.6-35B-A3B76.3 |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
HLE
KnowledgeA 52.1%B 21.4%Winner: GPT-5.4Δ 30.7HLE: GPT-5.4 scored 52.1%; Qwen3.6-35B-A3B scored 21.4%. GPT-5.4 wins this benchmark. - Source ↗
Terminal-Bench 2.0
AgenticA 75.1%B 51.5%Winner: GPT-5.4Δ 23.6Terminal-Bench 2.0: GPT-5.4 scored 75.1%; Qwen3.6-35B-A3B scored 51.5%. GPT-5.4 wins this benchmark. - Source ↗
SWE-bench Pro
CodingA 57.7%B 49.5%Winner: GPT-5.4Δ 8.2SWE-bench Pro: GPT-5.4 scored 57.7%; Qwen3.6-35B-A3B scored 49.5%. GPT-5.4 wins this benchmark. - Source ↗
GPQA
KnowledgeA 92.8%B 86%Winner: GPT-5.4Δ 6.8GPQA: GPT-5.4 scored 92.8%; Qwen3.6-35B-A3B scored 86%. GPT-5.4 wins this benchmark. - Source ↗
MMMU-Pro
MultimodalA 81.2%B 75.3%Winner: GPT-5.4Δ 5.9MMMU-Pro: GPT-5.4 scored 81.2%; Qwen3.6-35B-A3B scored 75.3%. GPT-5.4 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | GPT-5.4 | Qwen3.6-35B-A3B | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | GPT-5.4$2.5 input / $15 output | Qwen3.6-35B-A3BNot available | A complete price comparison is not available. |
| Generation speedtokens per second | GPT-5.474 tok/s | Qwen3.6-35B-A3BNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | GPT-5.4151.79 s | Qwen3.6-35B-A3BNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | GPT-5.41.05M | Qwen3.6-35B-A3B262K | GPT-5.4 lists the larger context window. |
Benchmark Deep Dive
AgenticGPT-5.4 wins23 benchmarks
| Benchmark | GPT-5.4 | Qwen3.6-35B-A3B | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 75.1% | 51.5% | GPT-5.4 leads |
| CyberGymSource | 79.0% | — | Not comparable |
| BrowseCompSource | 82.7% | — | Not comparable |
| OSWorld-VerifiedSource | 75% | — | Not comparable |
| MCP AtlasSource | 70.6% | 62.8% | GPT-5.4 leads |
| ToolathlonSource | 54.6% | 26.9% | GPT-5.4 leads |
| τ²-bench resultsSource | 98.9% | 95.3% | GPT-5.4 leads |
| Claw-EvalSource | 60.3% | 68.7% | Qwen3.6-35B-A3B leads |
| DeepSearchQASource | 73.6% | — | Not comparable |
| AA Agentic IndexSource | 41.1% | 21.4% | GPT-5.4 leads |
| APEX-Agents-AASource | 33.3% | — | Not comparable |
| GDPval-AASource | 44.7% | 27.4% | GPT-5.4 leads |
| GDPval-AASource | 1395 | 1049 | GPT-5.4 leads |
| Gert LabsSource | 64.89% | 42.65% | GPT-5.4 leads |
| ResearchClawBenchSource | 15.3% | — | Not comparable |
| JobBenchSource | 38.9% | — | Not comparable |
| ExploitGymSource | 6.0% | — | Not comparable |
| QwenClawBenchSource | — | 52.6% | Not comparable |
| QwenWebBenchSource | — | 1397 | Not comparable |
| τ³-bench resultsSource | — | 67.2% | Not comparable |
| VITA-BenchSource | — | 35.6% | Not comparable |
| DeepPlanningSource | — | 25.9% | Not comparable |
| WideResearchSource | — | 60.1% | Not comparable |
CodingQwen3.6-35B-A3B wins12 benchmarks
| Benchmark | GPT-5.4 | Qwen3.6-35B-A3B | Result |
|---|---|---|---|
| LiveCodeBench ProSource | 87.5% | — | Not comparable |
| SWE-bench ProSource | 57.7% | 49.5% | GPT-5.4 leads |
| React Native EvalsSource | 85.3% | — | Not comparable |
| Vibe Code BenchSource | 67.42% | — | Not comparable |
| AA Coding IndexSource | 71.0% | 41.9% | GPT-5.4 leads |
| Terminal-Bench HardSource | 57.6% | 34.8% | GPT-5.4 leads |
| AA-SciCodeSource | 56.6% | 35.8% | GPT-5.4 leads |
| SWE-bench VerifiedSource | — | 73.4% | Not comparable |
| SWE MultilingualSource | — | 67.2% | Not comparable |
| Terminal-Bench 2.0Source | — | 51.5% | Not comparable |
| LiveCodeBenchSource | — | 80.4% | Not comparable |
| NL2RepoSource | — | 29.4% | Not comparable |
Reasoning2 benchmarks
KnowledgeGPT-5.4 wins16 benchmarks
| Benchmark | GPT-5.4 | Qwen3.6-35B-A3B | Result |
|---|---|---|---|
| GPQASource | 92.8% | 86% | GPT-5.4 leads |
| HLESource | 52.1% | 21.4% | GPT-5.4 leads |
| HLE w/o toolsSource | 39.8% | — | Not comparable |
| GPQA-DSource | 92.8% | — | Not comparable |
| HealthBench HardSource | 40.1% | — | Not comparable |
| MedXpertQA (Text)Source | 59.6% | — | Not comparable |
| Artificial Analysis Intelligence IndexSource | 51.4% | 31.6% | GPT-5.4 leads |
| AA-GPQA DiamondSource | 92.0% | 84.1% | GPT-5.4 leads |
| AA-HLESource | 41.6% | 20.2% | GPT-5.4 leads |
| AA-Omniscience IndexSource | 5.7% | -21.4% | GPT-5.4 leads |
| AA-Omniscience AccuracySource | 50.0% | 18.9% | GPT-5.4 leads |
| AA-Omniscience Hallucination RateSource | 88.6% | 49.7% | Qwen3.6-35B-A3B leads |
| HealthBench ProfessionalSource | 48.1% | — | Not comparable |
| MMLU-ProSource | — | 85.2% | Not comparable |
| SuperGPQASource | — | 64.7% | Not comparable |
| C-EvalSource | — | 90% | Not comparable |
MathQwen3.6-35B-A3B wins7 benchmarks
| Benchmark | GPT-5.4 | Qwen3.6-35B-A3B | Result |
|---|---|---|---|
| FrontierMath v2 (Tiers 1-3)Source | 47.600% | — | Not comparable |
| FrontierMath v2 (Tier 4)Source | 27.100% | — | Not comparable |
| HMMT Feb 2025Source | — | 90.7% | Not comparable |
| HMMT Nov 2025Source | — | 89.1% | Not comparable |
| HMMT Feb 2026Source | — | 83.6% | Not comparable |
| MMAnswerBenchSource | — | 78.9% | Not comparable |
| AIME26Source | — | 92.7% | Not comparable |
MultimodalQwen3.6-35B-A3B wins23 benchmarks
| Benchmark | GPT-5.4 | Qwen3.6-35B-A3B | Result |
|---|---|---|---|
| MMMU-ProSource | 81.2% | 75.3% | GPT-5.4 leads |
| OfficeQA ProSource | 53.2% | — | Not comparable |
| MMMU-Pro w/ PythonSource | 82.1% | — | Not comparable |
| CharXivSource | 82.8% | 78% | GPT-5.4 leads |
| ERQASource | 65.4% | — | Not comparable |
| SimpleVQASource | 61.1% | 58.9% | GPT-5.4 leads |
| ScreenSpot ProSource | 85.4% | — | Not comparable |
| ZeroBenchSource | 41.0% | — | Not comparable |
| MedXpertQA (MM)Source | 77.1% | — | Not comparable |
| GDPval-AASource | 1672 | — | Not comparable |
| AA-MMMU-ProSource | 78.4% | 75.0% | GPT-5.4 leads |
| Design Arena WebsiteSource | 1254 | — | Not comparable |
| MMMUSource | — | 81.7% | Not comparable |
| RealWorldQASource | — | 85.3% | Not comparable |
| OmniDocBench 1.5Source | — | 89.9% | Not comparable |
| CC-OCRSource | — | 81.9% | Not comparable |
| AI2D_TESTSource | — | 92.7% | Not comparable |
| RefCOCO (avg)Source | — | 92.0% | Not comparable |
| ODINW13Source | — | 50.8% | Not comparable |
| Video-MME (with subtitle)Source | — | 86.6% | Not comparable |
| Video-MME (w/o subtitle)Source | — | 82.5% | Not comparable |
| VideoMMMUSource | — | 83.7% | Not comparable |
| MLVU (M-Avg)Source | — | 86.2% | Not comparable |
Inst. Following1 benchmarks
| Benchmark | GPT-5.4 | Qwen3.6-35B-A3B | Result |
|---|---|---|---|
| AA-IFBenchSource | 73.9% | 64.4% | GPT-5.4 leads |
Frequently Asked Questions (6)
Which is better, GPT-5.4 or Qwen3.6-35B-A3B?
GPT-5.4 is ahead on BenchLM's provisional leaderboard, 86 to 59. The biggest single separator in this matchup is HLE, where the scores are 52.1% and 21.4%.
Which is better for knowledge tasks, GPT-5.4 or Qwen3.6-35B-A3B?
GPT-5.4 has the edge for knowledge tasks in this comparison, averaging 57.6 versus 51.8. Inside this category, AA-Omniscience Hallucination Rate is the benchmark that creates the most daylight between them.
Which is better for coding, GPT-5.4 or Qwen3.6-35B-A3B?
Qwen3.6-35B-A3B has the edge for coding in this comparison, averaging 73.8 versus 57.7. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.
Which is better for math, GPT-5.4 or Qwen3.6-35B-A3B?
Qwen3.6-35B-A3B has the edge for math in this comparison, averaging 88.2 versus 42.5. GPT-5.4 stays close enough that the answer can still flip depending on your workload.
Which is better for agentic tasks, GPT-5.4 or Qwen3.6-35B-A3B?
GPT-5.4 has the edge for agentic tasks in this comparison, averaging 77.2 versus 51.5. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.
Which is better for multimodal and grounded tasks, GPT-5.4 or Qwen3.6-35B-A3B?
Qwen3.6-35B-A3B has the edge for multimodal and grounded tasks in this comparison, averaging 76.3 versus 73.2. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.
Related Comparisons
Explore More
The AI models change fast. We track them for you.
A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.