Model comparison
GPT-5.4 vs Qwen3 235B 2507 (Reasoning)
Head-to-head evidence from 2 shared benchmark results across 1 category. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: GPT-5.4 #8 (Supported); Qwen3 235B 2507 (Reasoning) #66 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. GPT-5.4 and Qwen3 235B 2507 (Reasoning) share 2 comparable benchmark results. 1 of 8 categories are comparable. 50 results are unique to GPT-5.4; 0 to Qwen3 235B 2507 (Reasoning).
Updated July 20, 2026- Shared results
- 2
- GPT-5.4 only
- 50
- Qwen3 235B 2507 (Reasoning) only
- 0
- Comparable categories
- 1 / 8
Pick GPT-5.4 if you want the stronger benchmark profile. Qwen3 235B 2507 (Reasoning) only becomes the better choice if you want the cheaper token bill.
Confidence note. This is a partial-evidence comparison with 2 shared benchmark results across 1 evidence category; 1 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 58.01. 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 6.4. The single biggest benchmark swing on the page is FrontierMath v2 (Tiers 1-3), 47.600% to 8.481%.
GPT-5.4 is also the more expensive model on tokens at $2.50 input / $15.00 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Qwen3 235B 2507 (Reasoning). That is roughly Infinityx on output cost alone. GPT-5.4 gives you the larger context window at 1.05M, compared with 128K for Qwen3 235B 2507 (Reasoning).
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 235B 2507 (Reasoning) |
|---|---|---|---|
| Math | GPT-5.442.5 | Margin← 36.1 | Qwen3 235B 2507 (Reasoning)6.4 |
| Agentic | GPT-5.477.2 | MarginNo overlap | Qwen3 235B 2507 (Reasoning)Not measured |
| Coding | GPT-5.457.7 | MarginNo overlap | Qwen3 235B 2507 (Reasoning)Not measured |
| Knowledge | GPT-5.457.6 | MarginNo overlap | Qwen3 235B 2507 (Reasoning)Not measured |
| Multimodal | GPT-5.473.2 | MarginNo overlap | Qwen3 235B 2507 (Reasoning)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 8.481%Winner: GPT-5.4Δ 39.1FrontierMath v2 (Tiers 1-3): GPT-5.4 scored 47.600%; Qwen3 235B 2507 (Reasoning) scored 8.481%. GPT-5.4 wins this benchmark. - Source ↗
FrontierMath v2 (Tier 4)
MathA 27.100%B 0.000%Winner: GPT-5.4Δ 27.1FrontierMath v2 (Tier 4): GPT-5.4 scored 27.100%; Qwen3 235B 2507 (Reasoning) scored 0.000%. 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 235B 2507 (Reasoning) | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | GPT-5.4$2.5 input / $15 output | Qwen3 235B 2507 (Reasoning)$0 input / $0 output | Qwen3 235B 2507 (Reasoning) has the lower combined listed price. |
| Generation speedtokens per second | GPT-5.474 tok/s | Qwen3 235B 2507 (Reasoning)Not available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | GPT-5.4151.79 s | Qwen3 235B 2507 (Reasoning)Not available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | GPT-5.41.05M | Qwen3 235B 2507 (Reasoning)128K | GPT-5.4 lists the larger context window. |
Benchmark Deep Dive
Agentic17 benchmarks
| Benchmark | GPT-5.4 | Qwen3 235B 2507 (Reasoning) | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 75.1% | — | Not comparable |
| 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% | — | Not comparable |
| 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 |
Coding6 benchmarks
| Benchmark | GPT-5.4 | Qwen3 235B 2507 (Reasoning) | Result |
|---|---|---|---|
| LiveCodeBench ProSource | 87.5% | — | Not comparable |
| SWE-bench ProSource | 57.7% | — | Not comparable |
| 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% | — | Not comparable |
Reasoning2 benchmarks
Knowledge13 benchmarks
| Benchmark | GPT-5.4 | Qwen3 235B 2507 (Reasoning) | 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% | — | Not comparable |
| AA-GPQA DiamondSource | 92.0% | — | Not comparable |
| AA-HLESource | 41.6% | — | Not comparable |
| AA-Omniscience IndexSource | 5.7% | — | Not comparable |
| AA-Omniscience AccuracySource | 50.0% | — | Not comparable |
| AA-Omniscience Hallucination RateSource | 88.6% | — | Not comparable |
| HealthBench ProfessionalSource | 48.1% | — | Not comparable |
MathGPT-5.4 wins2 benchmarks
Multimodal11 benchmarks
| Benchmark | GPT-5.4 | Qwen3 235B 2507 (Reasoning) | 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 | 1252 | — | Not comparable |
Inst. Following1 benchmarks
| Benchmark | GPT-5.4 | Qwen3 235B 2507 (Reasoning) | Result |
|---|---|---|---|
| AA-IFBenchSource | 73.9% | — | Not comparable |
Frequently Asked Questions (2)
Which is better, GPT-5.4 or Qwen3 235B 2507 (Reasoning)?
GPT-5.4 is ahead on BenchLM's BenchAlign leaderboard, 74.24 to 58.01. The biggest single separator in this matchup is FrontierMath v2 (Tiers 1-3), where the scores are 47.600% and 8.481%.
Which is better for math, GPT-5.4 or Qwen3 235B 2507 (Reasoning)?
GPT-5.4 has the edge for math in this comparison, averaging 42.5 versus 6.4. Inside this category, FrontierMath v2 (Tiers 1-3) 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.