LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GPT-5.4 Mini vs Qwen3.6 Plus

Updated October 21, 2026. 69 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GPT-5.4 Mini Replace Qwen3.6 Plus
GPT-5.4 MiniOpenAI
Coding
28.2
Overall
24.1
Output $/M
$4.50
18
Qwen3.6 PlusQwen
Coding
37.4
Overall
27.0
Output $/M
$1.95
50
vs boards won of 69 shared, 1 tied

Decision reading

Pick Qwen3.6 Plus for coding: it scores 37.4 on the Vibe Coding Index against 28.2 for GPT-5.4 Mini.

Pick GPT-5.4 Mini when a solved task costs $0.70 against $1.19, or when what matters is AA Coding Index or Vibe Code Bench (Vals AI), where it leads.

Model AGPT-5.4 MiniOpenAI28.2/1001 independent benchmark$0.70 per solved taskModel BQwen3.6 PlusQwen37.4/1001 independent benchmark$1.19 per solved task
Shared results
69
GPT-5.4 Mini only
19
Qwen3.6 Plus only
14
Like-for-like categories
7 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
GPT-5.4 Mini 28.2 from 1 benchmark, Qwen3.6 Plus 37.4 from 1 benchmark.
Widest category
Agentic work, 7.0 points apart on average over 5 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): GPT-5.4 Mini 48.0% against Qwen3.6 Plus 25.6%, 22.4 points apart on a 0 to 100 scale.
Output price
GPT-5.4 Mini $4.50 against Qwen3.6 Plus $1.95 per million output tokens, 2.3x apart.
Context window
GPT-5.4 Mini takes 400K tokens, Qwen3.6 Plus 1M.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryGPT-5.4 MiniBarsQwen3.6 PlusTies
Coding671 tie
Intelligence11
Agentic work32
Reasoning28
Math23
Knowledge19
General320
All shared boards18501 tie
CodingLike-for-likeFour of these boards feed the Vibe Coding Index67
Coding benchmarks for GPT-5.4 Mini and Qwen3.6 Plus
BenchmarkGPT-5.4 MiniQwen3.6 PlusLeader
LMArena WebDev13971461Qwen3.6 Plus
AA Coding Index56.154.5GPT-5.4 Mini
LiveCodeBench (Vals AI)81.5%86.0%Qwen3.6 Plus
SWE-bench Verified (Vals AI)73.0%73.4%Qwen3.6 Plus
Terminal-Bench 4 (Vals AI)2.5%Not measuredOnly GPT-5.4 Mini measured
Vibe Code Bench (Vals AI)48.0%25.6%GPT-5.4 Mini
Code migration (Vals AI)12.9%11.1%GPT-5.4 Mini
ProgramBench (Vals AI)0.0%0.0%Tie
Terminal-Bench 2.1 (Vals AI)54.7%53.2%GPT-5.4 Mini
SciCode52.1%Not measuredOnly GPT-5.4 Mini measured
Terminal-Bench Hard (Artificial Analysis run)52.3%43.9%GPT-5.4 Mini
FrontierCode27.0%Not measuredOnly GPT-5.4 Mini measured
WeirdML60.3%Not measuredOnly GPT-5.4 Mini measured
ALE-Bench1188.6670.1GPT-5.4 Mini
LMArena Text: coding14381467Qwen3.6 Plus
LMArena Text: software and IT14361462Qwen3.6 Plus
LMArena WebDev: React13861454Qwen3.6 Plus
LMArena WebDev: HTML14191463Qwen3.6 Plus
LMArena WebDev: from an imageNot measured1472Only Qwen3.6 Plus measured
Neither model has a result on the other 19 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
Intelligence benchmarks for GPT-5.4 Mini and Qwen3.6 Plus
BenchmarkGPT-5.4 MiniQwen3.6 PlusLeader
AA Intelligence Index24.127.0Qwen3.6 Plus
Epoch Capabilities Index148.7147.3GPT-5.4 Mini
Agentic workLike-for-likePublic boards, shown as published, kept out of the index32
Agentic work benchmarks for GPT-5.4 Mini and Qwen3.6 Plus
BenchmarkGPT-5.4 MiniQwen3.6 PlusLeader
AA Agentic Index17.9Not measuredOnly GPT-5.4 Mini measured
Finance Agent (Vals AI)45.4%40.9%GPT-5.4 Mini
Tax Agent Bench (Vals AI)8.8%7.7%GPT-5.4 Mini
Legal Research Bench (Vals AI)12.5%14.9%Qwen3.6 Plus
Excel Modeling Benchmark (Vals AI)45.4%32.9%GPT-5.4 Mini
τ²-Bench Telecom83.3%97.7%Qwen3.6 Plus
DeepResearch Bench36.3%Not measuredOnly GPT-5.4 Mini measured
Vending-Bench 2Not measured$5,115Only Qwen3.6 Plus measured
Neither model has a result on the other 31 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index28
Reasoning benchmarks for GPT-5.4 Mini and Qwen3.6 Plus
BenchmarkGPT-5.4 MiniQwen3.6 PlusLeader
GPQA Diamond83.1%87.4%Qwen3.6 Plus
MMMU Pro (Vals AI)79.3%84.2%Qwen3.6 Plus
GPQA Diamond (Artificial Analysis run)87.5%88.2%Qwen3.6 Plus
Humanity's Last Exam (Artificial Analysis run)28.1%27.8%GPT-5.4 Mini
GPQA Diamond (Epoch AI run)86.9%88.4%Qwen3.6 Plus
ARC-AGI63.7%Not measuredOnly GPT-5.4 Mini measured
ARC-AGI-218.9%Not measuredOnly GPT-5.4 Mini measured
Chess puzzles23.0%14.0%GPT-5.4 Mini
Mystery game puzzles11.0%12.0%Qwen3.6 Plus
LMArena Text: hard prompts14241448Qwen3.6 Plus
LMArena Text: expert14351454Qwen3.6 Plus
LMArena Text: hard prompts in English14201455Qwen3.6 Plus
LMArena Vision: diagrams1266Not measuredOnly GPT-5.4 Mini measured
LMArena Vision: homework1285Not measuredOnly GPT-5.4 Mini measured
Neither model has a result on the other 7 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index23
Math benchmarks for GPT-5.4 Mini and Qwen3.6 Plus
BenchmarkGPT-5.4 MiniQwen3.6 PlusLeader
AIME95.6%94.6%GPT-5.4 Mini
OTIS Mock AIME88.9%93.3%Qwen3.6 Plus
FrontierMath, tiers 1 to 351.2%38.2%GPT-5.4 Mini
FrontierMath, tier 49.8%Not measuredOnly GPT-5.4 Mini measured
ProofBench21.0%Not measuredOnly GPT-5.4 Mini measured
LMArena Text: math14181449Qwen3.6 Plus
LMArena Text: mathematical work14291448Qwen3.6 Plus
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index19
Knowledge benchmarks for GPT-5.4 Mini and Qwen3.6 Plus
BenchmarkGPT-5.4 MiniQwen3.6 PlusLeader
MMLU Pro84.5%87.7%Qwen3.6 Plus
LegalBench (Vals AI)Not measured84.2%Only Qwen3.6 Plus measured
CaseLaw (Vals AI)51.7%51.5%GPT-5.4 Mini
CorpFin (Vals AI)60.9%61.9%Qwen3.6 Plus
TaxEval (Vals AI)71.2%74.7%Qwen3.6 Plus
MortgageTax (Vals AI)63.5%68.0%Qwen3.6 Plus
MedCode (Vals AI)Not measured36.9%Only Qwen3.6 Plus measured
SimpleQA Verified29.4%44.1%Qwen3.6 Plus
LMArena Text: science14171439Qwen3.6 Plus
LMArena Text: medicine and healthcare13951442Qwen3.6 Plus
LMArena Text: legal and government14141437Qwen3.6 Plus
LMArena Text: business and finance14181430Qwen3.6 Plus
Neither model has a result on the other 5 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
Web design benchmarks for GPT-5.4 Mini and Qwen3.6 Plus
BenchmarkGPT-5.4 MiniQwen3.6 PlusLeader
Design Arena: websitesNot measured1243Only Qwen3.6 Plus measured
Design Arena: UI componentsNot measured1233Only Qwen3.6 Plus measured
Design Arena: game devNot measured1211Only Qwen3.6 Plus measured
Design Arena: data vizNot measured1228Only Qwen3.6 Plus measured
Design Arena: 3DNot measured1202Only Qwen3.6 Plus measured
Design Arena: all code categoriesNot measured1236Only Qwen3.6 Plus measured
Design Arena: SVGNot measured1160Only Qwen3.6 Plus measured
Design Arena: ASCII artNot measured1115Only Qwen3.6 Plus measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index320
General benchmarks for GPT-5.4 Mini and Qwen3.6 Plus
BenchmarkGPT-5.4 MiniQwen3.6 PlusLeader
SAGE (Vals AI)50.8%44.9%GPT-5.4 Mini
MedScribe (Vals AI)Not measured77.0%Only Qwen3.6 Plus measured
Vals Multimodal Index54.2%51.5%GPT-5.4 Mini
IFBench73.3%75.2%Qwen3.6 Plus
Long context reasoning (AA-LCR)77.0%78.3%Qwen3.6 Plus
DTBench80.0%81.9%Qwen3.6 Plus
LMCA40.8%33.1%GPT-5.4 Mini
CL-benchNot measured20.3%Only Qwen3.6 Plus measured
LMArena Text14121436Qwen3.6 Plus
LMArena Text: instruction following14051424Qwen3.6 Plus
LMArena Text: creative writing13701404Qwen3.6 Plus
LMArena Vision1245Not measuredOnly GPT-5.4 Mini measured
LMArena Text: English14121445Qwen3.6 Plus
LMArena Text: non-English14051424Qwen3.6 Plus
LMArena Text: Chinese14461477Qwen3.6 Plus
LMArena Text: German14081451Qwen3.6 Plus
LMArena Text: Japanese13751390Qwen3.6 Plus
LMArena Text: Korean13701379Qwen3.6 Plus
LMArena Text: French14411455Qwen3.6 Plus
LMArena Text: Spanish14051432Qwen3.6 Plus
LMArena Text: Russian14171434Qwen3.6 Plus
LMArena Text: Polish14151433Qwen3.6 Plus
LMArena Text: multi-turn14291438Qwen3.6 Plus
LMArena Text: longer queries14081439Qwen3.6 Plus
LMArena Text: writing and language13931418Qwen3.6 Plus
LMArena Text: entertainment and media13731402Qwen3.6 Plus
LMArena Vision: English1244Not measuredOnly GPT-5.4 Mini measured
LMArena Vision: Chinese1286Not measuredOnly GPT-5.4 Mini measured
LMArena Vision: reading text1253Not measuredOnly GPT-5.4 Mini measured
LMArena Vision: humor1235Not measuredOnly GPT-5.4 Mini measured
LMArena Vision: recognizing things1218Not measuredOnly GPT-5.4 Mini measured
LMArena Vision: creative writing1222Not measuredOnly GPT-5.4 Mini measured
Neither model has a result on the other 14 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-5.4 Mini and Qwen3.6 Plus
BenchmarkGPT-5.4 MiniQwen3.6 PlusLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

GPT-5.4 Mini$0.0030Fits in one request
Qwen3.6 Plus$0.0013Fits in one request

Qwen3.6 Plus costs less

Repository review

50K input + 3K output tokens

GPT-5.4 Mini$0.051Fits in one request
Qwen3.6 Plus$0.022Fits in one request

Qwen3.6 Plus costs less

Agent loop

200K input + 10K output tokens

GPT-5.4 Mini$0.195Fits in one request
Qwen3.6 Plus$0.085Fits in one request

Qwen3.6 Plus costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of GPT-5.4 Mini and Qwen3.6 Plus
SpecificationOpenAI GPT-5.4 MiniQwen3.6 Plus
MakerOpenAIQwen
WeightsNo public weights listedNo public weights listed
ReleasedMarch 17, 2026April 2, 2026
Context window400K1M
Input $/Mtok$0.75$0.325
Output $/Mtok$4.50$1.95
Coding index28.237.4
Independent measurements behind it11
Cost per solved task$0.70$1.19
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5.4 Mini better than Qwen3.6 Plus for coding?

Pick Qwen3.6 Plus for coding: it scores 37.4 on the Vibe Coding Index against 28.2 for GPT-5.4 Mini. Pick GPT-5.4 Mini when a solved task costs $0.70 against $1.19, or when what matters is AA Coding Index or Vibe Code Bench (Vals AI), where it leads.

Which is cheaper, GPT-5.4 Mini or Qwen3.6 Plus?

GPT-5.4 Mini costs $0.75 per million input tokens and $4.50 per million output tokens; Qwen3.6 Plus costs $0.325 and $1.95. Per solved SWE-bench Verified task, GPT-5.4 Mini costs $0.70 and Qwen3.6 Plus $1.19.

Which has the bigger context window, GPT-5.4 Mini or Qwen3.6 Plus?

Qwen3.6 Plus, with 1M tokens against 400K.

How many benchmarks is the GPT-5.4 Mini vs Qwen3.6 Plus comparison based on?

69 benchmarks have a result for both models as of October 21, 2026. GPT-5.4 Mini's coding index rests on 1 independent benchmark and Qwen3.6 Plus's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next