Math
Like-for-like- GPT-5.4 nano
- 21.0
- Qwen 3.6 Max (preview)
- 18.4
- Weighted basis
- 2 vs 2 rows
- Reading
- GPT-5.4 nano leads
Model comparison
GPT-5.4 nano vs Qwen 3.6 Max (preview)
Updated August 11, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
GPT-5.4 nano has the higher public score estimate, 66.47 versus 59.65, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
3 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.
Which one for your work
Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.
Long documents
Prompts that approach the documented context limit
GPT-5.4 nano
GPT-5.4 nano has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
No shared weighted benchmark basis supports a winner.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
10 GPT-5.4 nano only
3 Shared
6 Qwen 3.6 Max (preview) only
- Shared results
- 3
- GPT-5.4 nano only
- 10
- Qwen 3.6 Max (preview) only
- 6
- Like-for-like categories
- 1 / 8
1 category uses different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.
Agentic
Directional only- GPT-5.4 nano
- 42.9
- Qwen 3.6 Max (preview)
- 65.4
- Weighted basis
- 2 vs 1 rows
- Reading
- Directional only
Coding
Not comparable- GPT-5.4 nano
- Not measured
- Qwen 3.6 Max (preview)
- 51.0
- Weighted basis
- 0 vs 2 rows
- Reading
- Not comparable
Reasoning
Not comparable- GPT-5.4 nano
- Not measured
- Qwen 3.6 Max (preview)
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Knowledge
Not comparable- GPT-5.4 nano
- 43.8
- Qwen 3.6 Max (preview)
- 73.9
- Weighted basis
- 2 vs 1 rows
- Reading
- Not comparable
Multilingual
Not comparable- GPT-5.4 nano
- Not measured
- Qwen 3.6 Max (preview)
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Multimodal
Not comparable- GPT-5.4 nano
- 66.1
- Qwen 3.6 Max (preview)
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Instruction following
Not comparable- GPT-5.4 nano
- Not measured
- Qwen 3.6 Max (preview)
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
| Category | GPT-5.4 nano | Qwen 3.6 Max (preview) | Weighted basis | Reading |
|---|---|---|---|---|
| Math | 21.0 | 18.4 | Like-for-like2 vs 2 rows | GPT-5.4 nano leads |
| Agentic | 42.9 | 65.4 | Directional only2 vs 1 rows | Directional only |
| Coding | Not measured | 51.0 | Not comparable0 vs 2 rows | Not comparable |
| Reasoning | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Knowledge | 43.8 | 73.9 | Not comparable2 vs 1 rows | Not comparable |
| Multilingual | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Multimodal | 66.1 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Instruction following | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
Shape of the matched evidence
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.
Terminal-Bench 2.0
Agentic
GPT-5.4 nano: 46.3%Qwen 3.6 Max (preview): 65.4%Normalized gap 19.1GPT-5.4 nano source Qwen 3.6 Max (preview) sourceFrontierMath v2 (Tiers 1-3)
Math
FrontierMath v2 (Tier 4)
Math
What each workload costs
Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.
Chat turn
1K fresh input + 500 output tokens
- GPT-5.4 nano
- $0.00082
- Fits in one request
- Qwen 3.6 Max (preview)
- API rate not published
- Fits in one request
Qwen 3.6 Max (preview) has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- GPT-5.4 nano
- $0.01375
- Fits in one request
- Qwen 3.6 Max (preview)
- API rate not published
- Fits in one request
Qwen 3.6 Max (preview) has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- GPT-5.4 nano
- $0.0205
- Fits in one request
- Qwen 3.6 Max (preview)
- API rate not published
- Fits in one request
- Cached-input rate unavailable
Qwen 3.6 Max (preview) has no comparable published API token rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationGPT-5.4 nanoQwen 3.6 Max (preview)
Context window
Maximum documented context; output-token limits may be lower.
GPT-5.4 nano
Qwen 3.6 Max (preview)
256K
API model ID
GPT-5.4 nano
gpt-5.4-nano
OpenAI GPT-5.4 nano model documentationQwen 3.6 Max (preview)
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
GPT-5.4 nano
$0.02 per 1M cached input tokens
OpenAI pricingQwen 3.6 Max (preview)
No comparable hosted API rate
Documented inputs
GPT-5.4 nano
text, image
OpenAI model catalogQwen 3.6 Max (preview)
Not sourced
Documented outputs
GPT-5.4 nano
Qwen 3.6 Max (preview)
Not sourced
Provider availability
GPT-5.4 nano
Generally Available · OpenAI Responses API
OpenAI model catalogQwen 3.6 Max (preview)
Not sourced
Reasoning profile
GPT-5.4 nano
Reasoning
Qwen 3.6 Max (preview)
Reasoning
Weight access
GPT-5.4 nano
Proprietary
Qwen 3.6 Max (preview)
Proprietary
License
GPT-5.4 nano
Proprietary
Qwen 3.6 Max (preview)
Proprietary
Release date
GPT-5.4 nano
2026-03-17
Qwen 3.6 Max (preview)
2026-04-20
If you already use one of these models
- Deployment change
- The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
- Quality signal
- GPT-5.4 nano has the higher public score estimate, 66.47 versus 59.65, but the 90% score intervals overlap.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- GPT-5.4 nano has the larger documented window (400K).
Run the same representative tasks against both endpoints before changing production traffic.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence19 rows
Agentic
Terminal-Bench 2.0
GPT-5.4 nano46.3%SourceQwen 3.6 Max (preview)65.4%SourceQwen 3.6 Max (preview) leads this result
OSWorld-Verified
GPT-5.4 nano39%SourceQwen 3.6 Max (preview)—Not directly comparable
MCP Atlas
GPT-5.4 nano56.1%SourceQwen 3.6 Max (preview)—Not directly comparable
Toolathlon
GPT-5.4 nano35.5%SourceQwen 3.6 Max (preview)—Not directly comparable
τ²-bench results
GPT-5.4 nano92.5%SourceQwen 3.6 Max (preview)—Not directly comparable
QwenClawBench
GPT-5.4 nano—Qwen 3.6 Max (preview)59.0%SourceNot directly comparable
Coding
Vibe Code Bench
GPT-5.4 nano26.10%SourceQwen 3.6 Max (preview)—Not directly comparable
SWE-bench Pro
GPT-5.4 nano—Qwen 3.6 Max (preview)57.3%SourceNot directly comparable
SciCode
GPT-5.4 nano—Qwen 3.6 Max (preview)47%SourceNot directly comparable
NL2Repo
GPT-5.4 nano—Qwen 3.6 Max (preview)42.9%SourceNot directly comparable
Terminal-Bench 2.0
GPT-5.4 nano—Qwen 3.6 Max (preview)65.4%SourceNot directly comparable
Knowledge
GPQA
GPT-5.4 nano82.8%SourceQwen 3.6 Max (preview)—Not directly comparable
HLE
GPT-5.4 nano37.7%SourceQwen 3.6 Max (preview)—Not directly comparable
HLE w/o tools
GPT-5.4 nano24.3%SourceQwen 3.6 Max (preview)—Not directly comparable
SuperGPQA
GPT-5.4 nano—Qwen 3.6 Max (preview)73.9%SourceNot directly comparable
Math
FrontierMath v2 (Tiers 1-3)
Shared sourceGPT-5.4 nano25.860%Qwen 3.6 Max (preview)23.103%GPT-5.4 nano leads this result
FrontierMath v2 (Tier 4)
Shared sourceGPT-5.4 nano6.250%Qwen 3.6 Max (preview)4.167%GPT-5.4 nano leads this result
Multimodal
Frequently asked questions
Which is better, GPT-5.4 nano or Qwen 3.6 Max (preview)?
GPT-5.4 nano has the higher public score estimate, 66.47 versus 59.65, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, GPT-5.4 nano or Qwen 3.6 Max (preview)?
The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.
Which is better for agentic tasks, GPT-5.4 nano or Qwen 3.6 Max (preview)?
The current agentic tasks averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.
Which costs less, GPT-5.4 nano or Qwen 3.6 Max (preview)?
Both models do not have comparable published API token rates, so this page does not name a universal price winner.
Which has the larger context window, GPT-5.4 nano or Qwen 3.6 Max (preview)?
GPT-5.4 nano has the larger documented context window: 400K, compared with 256K.
Related comparisons
Last updated August 11, 2026
Watch GPT-5.4 nano vs Qwen 3.6 Max (preview)
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.