Model comparison
GPT-5.4 vs Qwen 3.6 Max (preview)
Head-to-head evidence from 15 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: GPT-5.4 #8 (Supported); Qwen 3.6 Max (preview) #53 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. GPT-5.4 and Qwen 3.6 Max (preview) share 15 comparable benchmark results. 4 of 8 categories are comparable. 37 results are unique to GPT-5.4; 6 to Qwen 3.6 Max (preview).
Updated July 23, 2026- Shared results
- 15
- GPT-5.4 only
- 37
- Qwen 3.6 Max (preview) only
- 6
- Comparable categories
- 4 / 8
Pick GPT-5.4 if you want the stronger benchmark profile. Qwen 3.6 Max (preview) only becomes the better choice if knowledge is the priority.
Confidence note. This is a partial-evidence comparison with 15 shared benchmark results across 6 evidence categories; 4 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 BenchAlign aggregate, 74.24 to 59.72. 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 mathematics, where it averages 42.5 against 18.4. The single biggest benchmark swing on the page is FrontierMath v2 (Tiers 1-3), 47.600% to 23.103%. Qwen 3.6 Max (preview) does hit back in knowledge, 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 256K for Qwen 3.6 Max (preview).
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 | Qwen 3.6 Max (preview) |
|---|---|---|---|
| Math | GPT-5.442.5 | Margin← 24.1 | Qwen 3.6 Max (preview)18.4 |
| Knowledge | GPT-5.457.6 | Margin→ 16.3 | Qwen 3.6 Max (preview)73.9 |
| Agentic | GPT-5.477.2 | Margin← 11.8 | Qwen 3.6 Max (preview)65.4 |
| Coding | GPT-5.457.7 | Margin← 6.7 | Qwen 3.6 Max (preview)51.0 |
| Multimodal | GPT-5.473.2 | MarginNo overlap | Qwen 3.6 Max (preview)Not measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
FrontierMath v2 (Tiers 1-3)
MathA 47.600%B 23.103%Winner: GPT-5.4Δ 24.5FrontierMath v2 (Tiers 1-3): GPT-5.4 scored 47.600%; Qwen 3.6 Max (preview) scored 23.103%. GPT-5.4 wins this benchmark. - Source ↗
FrontierMath v2 (Tier 4)
MathA 27.100%B 4.167%Winner: GPT-5.4Δ 22.9FrontierMath v2 (Tier 4): GPT-5.4 scored 27.100%; Qwen 3.6 Max (preview) scored 4.167%. GPT-5.4 wins this benchmark. - Source ↗
Terminal-Bench 2.0
AgenticA 75.1%B 65.4%Winner: GPT-5.4Δ 9.7Terminal-Bench 2.0: GPT-5.4 scored 75.1%; Qwen 3.6 Max (preview) scored 65.4%. GPT-5.4 wins this benchmark. - Source ↗
SWE-bench Pro
CodingA 57.7%B 57.3%Winner: GPT-5.4Δ 0.4SWE-bench Pro: GPT-5.4 scored 57.7%; Qwen 3.6 Max (preview) scored 57.3%. 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 | Qwen 3.6 Max (preview) | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | GPT-5.4$2.5 input / $15 output | Qwen 3.6 Max (preview)Not available | A complete price comparison is not available. |
| Generation speedtokens per second | GPT-5.474 tok/s | Qwen 3.6 Max (preview)Not available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | GPT-5.4151.79 s | Qwen 3.6 Max (preview)Not available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | GPT-5.41.05M | Qwen 3.6 Max (preview)256K | GPT-5.4 lists the larger context window. |
Benchmark Deep Dive
AgenticGPT-5.4 wins19 benchmarks
| Benchmark | GPT-5.4 | Qwen 3.6 Max (preview) | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 75.1% | 65.4% | GPT-5.4 leads |
| CyberGymSource | 79.0% | — | Not comparable |
| BrowseCompSource | 82.7% | — | Not comparable |
| OSWorld-VerifiedSource | 75% | — | Not comparable |
| MCP AtlasSource | 70.6% | — | Not comparable |
| ToolathlonSource | 54.6% | — | Not comparable |
| τ²-bench resultsSource | 87.1% | 95.9% | Qwen 3.6 Max (preview) leads |
| Claw-EvalSource | 60.3% | — | Not comparable |
| DeepSearchQASource | 73.6% | — | Not comparable |
| AA Agentic IndexSource | 41.1% | — | Not comparable |
| APEX-Agents-AASource | 33.3% | — | Not comparable |
| GDPval-AASource | 44.7% | — | Not comparable |
| GDPval-AASource | 1395 | — | Not comparable |
| Gert LabsSource | 64.89% | — | Not comparable |
| ResearchClawBenchSource | 15.3% | — | Not comparable |
| JobBenchSource | 38.9% | — | Not comparable |
| ExploitGymSource | 6.0% | — | Not comparable |
| QwenClawBenchSource | — | 59.0% | Not comparable |
| QwenWebBenchSource | — | 1532 | Not comparable |
CodingGPT-5.4 wins9 benchmarks
| Benchmark | GPT-5.4 | Qwen 3.6 Max (preview) | Result |
|---|---|---|---|
| LiveCodeBench ProSource | 87.5% | — | Not comparable |
| SWE-bench ProSource | 57.7% | 57.3% | GPT-5.4 leads |
| React Native EvalsSource | 85.3% | — | Not comparable |
| Vibe Code BenchSource | 67.42% | — | Not comparable |
| AA Coding IndexSource | 71.0% | — | Not comparable |
| AA-SciCodeSource | 56.6% | 46.9% | GPT-5.4 leads |
| SciCodeSource | — | 47% | Not comparable |
| NL2RepoSource | — | 42.9% | Not comparable |
| Terminal-Bench 2.0Source | — | 65.4% | Not comparable |
Reasoning2 benchmarks
KnowledgeQwen 3.6 Max (preview) wins14 benchmarks
| Benchmark | GPT-5.4 | Qwen 3.6 Max (preview) | Result |
|---|---|---|---|
| GPQASource | 92.8% | — | Not comparable |
| HLESource | 52.1% | — | Not comparable |
| 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% | 40.0% | GPT-5.4 leads |
| AA-GPQA DiamondSource | 92.0% | 88.8% | GPT-5.4 leads |
| AA-HLESource | 41.6% | 28.9% | GPT-5.4 leads |
| AA-Omniscience IndexSource | 5.7% | 10.2% | Qwen 3.6 Max (preview) leads |
| AA-Omniscience AccuracySource | 50.0% | 37.7% | GPT-5.4 leads |
| AA-Omniscience Hallucination RateSource | 88.6% | 44.2% | Qwen 3.6 Max (preview) leads |
| HealthBench ProfessionalSource | 48.1% | — | Not comparable |
| SuperGPQASource | — | 73.9% | Not comparable |
MathGPT-5.4 wins2 benchmarks
Multimodal11 benchmarks
| Benchmark | GPT-5.4 | Qwen 3.6 Max (preview) | Result |
|---|---|---|---|
| MMMU-ProSource | 81.2% | — | Not comparable |
| OfficeQA ProSource | 53.2% | — | Not comparable |
| MMMU-Pro w/ PythonSource | 82.1% | — | Not comparable |
| CharXivSource | 82.8% | — | Not comparable |
| ERQASource | 65.4% | — | Not comparable |
| SimpleVQASource | 61.1% | — | Not comparable |
| ScreenSpot ProSource | 85.4% | — | Not comparable |
| ZeroBenchSource | 41.0% | — | Not comparable |
| MedXpertQA (MM)Source | 77.1% | — | Not comparable |
| AA-MMMU-ProSource | 78.4% | — | Not comparable |
| Design Arena WebsiteSource | 1250 | — | Not comparable |
Inst. Following1 benchmarks
| Benchmark | GPT-5.4 | Qwen 3.6 Max (preview) | Result |
|---|---|---|---|
| AA-IFBenchSource | 73.9% | 76.6% | Qwen 3.6 Max (preview) leads |
Frequently Asked Questions (5)
Which is better, GPT-5.4 or Qwen 3.6 Max (preview)?
GPT-5.4 is ahead on BenchLM's BenchAlign leaderboard, 74.24 to 59.72. The biggest single separator in this matchup is FrontierMath v2 (Tiers 1-3), where the scores are 47.600% and 23.103%.
Which is better for knowledge tasks, GPT-5.4 or Qwen 3.6 Max (preview)?
Qwen 3.6 Max (preview) has the edge for knowledge tasks in this comparison, averaging 73.9 versus 57.6. 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 Qwen 3.6 Max (preview)?
GPT-5.4 has the edge for coding in this comparison, averaging 57.7 versus 51. Inside this category, AA-SciCode is the benchmark that creates the most daylight between them.
Which is better for math, GPT-5.4 or Qwen 3.6 Max (preview)?
GPT-5.4 has the edge for math in this comparison, averaging 42.5 versus 18.4. Inside this category, FrontierMath v2 (Tiers 1-3) is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, GPT-5.4 or Qwen 3.6 Max (preview)?
GPT-5.4 has the edge for agentic tasks in this comparison, averaging 77.2 versus 65.4. Inside this category, Terminal-Bench 2.0 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.