Skip to main content
BenchLM
Advertise

Model comparison

GPT-5.4 vs Qwen3.7 Max

Data verified

Head-to-head evidence from 25 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

GPT-5.4
OpenAI
74.24/100
Margin
1.4pts
← winning
72.84/100
1 category wins3 category wins

Public leaderboard positions: GPT-5.4 #8 (Supported); Qwen3.7 Max #10 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. GPT-5.4 and Qwen3.7 Max share 25 comparable benchmark results. 4 of 8 categories are comparable. 27 results are unique to GPT-5.4; 33 to Qwen3.7 Max.

Updated July 23, 2026
Shared results
25
GPT-5.4 only
27
Qwen3.7 Max only
33
Comparable categories
4 / 8

Pick GPT-5.4 if you want the stronger benchmark profile. Qwen3.7 Max only becomes the better choice if mathematics is the priority.

Confidence note. This is a partial-evidence comparison with 25 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 has the cleaner BenchAlign overall profile here, landing at 74.24 versus 72.84. It is a real lead, but still close enough that category-level strengths matter more than the headline number.

GPT-5.4's sharpest advantage is in agentic, where it averages 77.2 against 69.7. The single biggest benchmark swing on the page is HLE, 52.1% to 41.4%. Qwen3.7 Max 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 1M for Qwen3.7 Max.

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 GPT-5.4 and Qwen3.7 Max
CategoryGPT-5.4ΔQwen3.7 Max
MathGPT-5.442.5Margin→ 54.6Qwen3.7 Max97.1
CodingGPT-5.457.7Margin→ 20.2Qwen3.7 Max77.9
AgenticGPT-5.477.2Margin← 7.5Qwen3.7 Max69.7
KnowledgeGPT-5.457.6Margin→ 6.6Qwen3.7 Max64.2
ReasoningGPT-5.4Not measuredMarginNo overlapQwen3.7 Max90.4
MultilingualGPT-5.4Not measuredMarginNo overlapQwen3.7 Max87.0
MultimodalGPT-5.473.2MarginNo overlapQwen3.7 MaxNot measured
Inst. FollowingGPT-5.4Not measuredMarginNo overlapQwen3.7 Max84.4

Decisive benchmark drivers

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

More
A · GPT-5.4B · Qwen3.7 Max

  1. HLE

    Knowledge
    Source ↗
    A 52.1%B 41.4%
    Winner: GPT-5.4Δ 10.7
    HLE: GPT-5.4 scored 52.1%; Qwen3.7 Max scored 41.4%. GPT-5.4 wins this benchmark.

  2. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 75.1%B 69.7%
    Winner: GPT-5.4Δ 5.4
    Terminal-Bench 2.0: GPT-5.4 scored 75.1%; Qwen3.7 Max scored 69.7%. GPT-5.4 wins this benchmark.

  3. SWE-bench Pro

    Coding
    Source ↗
    A 57.7%B 60.6%
    Winner: Qwen3.7 MaxΔ 2.9
    SWE-bench Pro: GPT-5.4 scored 57.7%; Qwen3.7 Max scored 60.6%. Qwen3.7 Max wins this benchmark.

  4. GPQA

    Knowledge
    Source ↗
    A 92.8%B 92.4%
    Winner: GPT-5.4Δ 0.4
    GPQA: GPT-5.4 scored 92.8%; Qwen3.7 Max scored 92.4%. GPT-5.4 wins this benchmark.

Operational comparison

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

MetricGPT-5.4Qwen3.7 MaxComparison
Input / output priceUSD per 1M tokensGPT-5.4$2.5 input / $15 outputQwen3.7 MaxNot availableA complete price comparison is not available.
Generation speedtokens per secondGPT-5.474 tok/sQwen3.7 MaxNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenGPT-5.4151.79 sQwen3.7 MaxNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensGPT-5.41.05MQwen3.7 Max1MGPT-5.4 lists the larger context window.

Benchmark Deep Dive

AgenticGPT-5.4 wins
BenchmarkGPT-5.4Qwen3.7 MaxResult
Terminal-Bench 2.0Source 75.1%69.7%GPT-5.4 leads
CyberGymSource 79.0%—Not comparable
BrowseCompSource 82.7%—Not comparable
OSWorld-VerifiedSource 75%—Not comparable
MCP AtlasSource 70.6%76.4%Qwen3.7 Max leads
ToolathlonSource 54.6%—Not comparable
τ²-bench resultsSource 87.1%94.7%Qwen3.7 Max leads
Claw-EvalSource 60.3%65.2%Qwen3.7 Max leads
DeepSearchQASource 73.6%—Not comparable
AA Agentic IndexSource 41.1%30.6%GPT-5.4 leads
APEX-Agents-AASource 33.3%—Not comparable
GDPval-AASource 44.7%38.7%GPT-5.4 leads
GDPval-AASource 13951273GPT-5.4 leads
Gert LabsSource 64.89%64.27%GPT-5.4 leads
ResearchClawBenchSource 15.3%18.7%Qwen3.7 Max leads
JobBenchSource 38.9%—Not comparable
ExploitGymSource 6.0%—Not comparable
QwenClawBenchSource —64.3%Not comparable
QwenWebBenchSource —1568Not comparable
BFCL v4Source —75.0%Not comparable
VITA-BenchSource —47.9%Not comparable
HLE w/ toolsSource —53.5%Not comparable
AA BriefcaseSource —908Not comparable
AA AutomationBenchSource —25.6%Not comparable
AA EnterpriseOps-GymSource —45.0%Not comparable
AA ITBenchSource —42.5%Not comparable
terminalBenchHardSource —50.8%Not comparable
aaTerminalBench21Source —74.5%Not comparable
AA Harvey LABSource —83.4%Not comparable
CodingQwen3.7 Max wins
BenchmarkGPT-5.4Qwen3.7 MaxResult
LiveCodeBench ProSource 87.5%—Not comparable
SWE-bench ProSource 57.7%60.6%Qwen3.7 Max leads
React Native EvalsSource 85.3%—Not comparable
Vibe Code BenchSource 67.42%—Not comparable
AA Coding IndexSource 71.0%66.0%GPT-5.4 leads
AA-SciCodeSource 56.6%48.8%GPT-5.4 leads
SWE-bench VerifiedSource —80.4%Not comparable
SWE MultilingualSource —78.3%Not comparable
NL2RepoSource —47.2%Not comparable
SciCodeSource —53.5%Not comparable
LiveCodeBenchSource —91.6%Not comparable
Terminal-Bench 2.0Source —69.7%Not comparable
Reasoning
BenchmarkGPT-5.4Qwen3.7 MaxResult
AA-LCRSource 74.0%69.0%GPT-5.4 leads
CritPtSource 23.4%13.4%GPT-5.4 leads
MRCRv2Source —90.4%Not comparable
KnowledgeQwen3.7 Max wins
BenchmarkGPT-5.4Qwen3.7 MaxResult
GPQASource 92.8%92.4%GPT-5.4 leads
HLESource 52.1%41.4%GPT-5.4 leads
HLE w/o toolsSource 39.8%—Not comparable
GPQA-DSource 92.8%92.4%GPT-5.4 leads
HealthBench HardSource 40.1%—Not comparable
MedXpertQA (Text)Source 59.6%—Not comparable
Artificial Analysis Intelligence IndexSource 51.4%46.0%GPT-5.4 leads
AA-GPQA DiamondSource 92.0%92.3%Qwen3.7 Max leads
AA-HLESource 41.6%38.1%GPT-5.4 leads
AA-Omniscience IndexSource 5.7%14.1%Qwen3.7 Max leads
AA-Omniscience AccuracySource 50.0%30.1%GPT-5.4 leads
AA-Omniscience Hallucination RateSource 88.6%22.9%Qwen3.7 Max leads
HealthBench ProfessionalSource 48.1%—Not comparable
MMLU-ProSource —89.6%Not comparable
MMLU-ReduxSource —95%Not comparable
SuperGPQASource —73.6%Not comparable
MMMLUSource —90.3%Not comparable
MathQwen3.7 Max wins
BenchmarkGPT-5.4Qwen3.7 MaxResult
FrontierMath v2 (Tiers 1-3)Source 47.600%—Not comparable
FrontierMath v2 (Tier 4)Source 27.100%—Not comparable
HMMT Feb 2026Source —97.1%Not comparable
IMOAnswerBenchSource —90.0%Not comparable
ApexSource —44.5%Not comparable
Multilingual
BenchmarkGPT-5.4Qwen3.7 MaxResult
MMLU-ProXSource —87%Not comparable
NOVA-63Source —59.0%Not comparable
INCLUDESource —86.2%Not comparable
MAXIFESource —89.2%Not comparable
PolyMathSource —86.5%Not comparable
Multimodal
BenchmarkGPT-5.4Qwen3.7 MaxResult
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 12501293Qwen3.7 Max leads
Inst. Following
BenchmarkGPT-5.4Qwen3.7 MaxResult
AA-IFBenchSource 73.9%80.5%Qwen3.7 Max leads
IFEvalSource —94.3%Not comparable
IFBenchSource —79.1%Not comparable
Frequently Asked Questions (5)

Which is better, GPT-5.4 or Qwen3.7 Max?

GPT-5.4 is ahead on BenchLM's BenchAlign leaderboard, 74.24 to 72.84. The biggest single separator in this matchup is HLE, where the scores are 52.1% and 41.4%.

Which is better for knowledge tasks, GPT-5.4 or Qwen3.7 Max?

Qwen3.7 Max has the edge for knowledge tasks in this comparison, averaging 64.2 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 Qwen3.7 Max?

Qwen3.7 Max has the edge for coding in this comparison, averaging 77.9 versus 57.7. Inside this category, AA-SciCode is the benchmark that creates the most daylight between them.

Which is better for math, GPT-5.4 or Qwen3.7 Max?

Qwen3.7 Max has the edge for math in this comparison, averaging 97.1 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 Qwen3.7 Max?

GPT-5.4 has the edge for agentic tasks in this comparison, averaging 77.2 versus 69.7. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

OpenAI Compare Pricing Methodology Find Your Best LLM Overall Rankings
Last updated: July 23, 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.