Enter opens the highlighted result. Esc closes.

GPT-5.4 Mini vs GPT-6 Astra

GPT-6 Astra leads GPT-5.4 Mini on coding, 85.1 to 35.8 on the Vibe Coding Index, and GPT-5.4 Mini costs 11.1x less per output token ($4.50 vs $50.00 per million). Updated October 22, 2026; 71 benchmarks have a result for both.

Replace GPT-5.4 Mini Replace GPT-6 Astra
GPT-5.4 MiniOpenAI
Coding
35.8
Overall
44.0
Output $/M
$4.50
2
GPT-6 AstraOpenAI
Coding
85.1
Overall
83.0
Output $/M
$50.00
69
vs boards won of 71 shared

GPT-5.4 Mini or GPT-6 Astra: which to pick

Pick GPT-6 Astra for coding: it scores 85.1 on the Vibe Coding Index against 35.8 for GPT-5.4 Mini.

Pick GPT-5.4 Mini for output tokens 11.1x cheaper ($4.50 vs $50.00 per million) and work measured by SAGE (Vals AI) or LMArena Text: French, where it leads.

Model AGPT-5.4 MiniOpenAI35.8/1005 independent benchmarks$0.70 per solved taskModel BGPT-6 AstraOpenAI85.1/1005 independent benchmarksNo cost per solved task
Shared results
71
GPT-5.4 Mini only
18
GPT-6 Astra only
50
Like-for-like categories
7 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

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 35.8 from 5 benchmarks, GPT-6 Astra 85.1 from 5 benchmarks.
Widest category
Math, 39.4 points apart on average over 6 shared benchmarks.
Widest single gap
FrontierMath, tier 4: GPT-5.4 Mini 9.8% against GPT-6 Astra 97.6%, 87.8 points apart on a 0 to 100 scale.
Output price
GPT-5.4 Mini $4.50 against GPT-6 Astra $50.00 per million output tokens, 11.1x apart.
Context window
GPT-5.4 Mini takes 400K tokens, GPT-6 Astra 1.05M.

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 MiniBarsGPT-6 AstraTies
Coding015
Intelligence02
Agentic work05
Reasoning012
Math06
Knowledge05
General224
All shared boards269
CodingLike-for-likeSix of these boards feed the Vibe Coding Index015
Coding benchmarks for GPT-5.4 Mini and GPT-6 Astra
BenchmarkGPT-5.4 MiniGPT-6 AstraLeader
LiveBench codingNot measured93.7%Only GPT-6 Astra measured
LMArena WebDev13971774GPT-6 Astra
AA Coding Index56.176.9GPT-6 Astra
LiveCodeBench (Vals AI)81.5%Not measuredOnly GPT-5.4 Mini measured
SWE-bench Verified (Vals AI)73.0%Not measuredOnly GPT-5.4 Mini measured
Terminal-Bench 4 (Vals AI)2.5%59.6%GPT-6 Astra
Vibe Code Bench (Vals AI)48.0%89.6%GPT-6 Astra
Code migration (Vals AI)12.9%67.7%GPT-6 Astra
ProgramBench (Vals AI)0.0%5.5%GPT-6 Astra
IOI (Vals AI)Not measured100.0%Only GPT-6 Astra measured
Terminal-Bench 2.1 (Vals AI)54.7%87.3%GPT-6 Astra
SWE-bench Multimodal (Vals AI)Not measured43.3%Only GPT-6 Astra measured
Vibe Code Bench 1-100 (Vals AI)Not measured27.6%Only GPT-6 Astra measured
CyberBench (Vals AI)Not measured41.1%Only GPT-6 Astra measured
SciCode52.1%56.5%GPT-6 Astra
Terminal-Bench Hard (Artificial Analysis run)52.3%Not measuredOnly GPT-5.4 Mini measured
FrontierCode27.0%53.3%GPT-6 Astra
FrontierSWENot measured65.5%Only GPT-6 Astra measured
GSO-BenchNot measured79.4%Only GPT-6 Astra measured
WeirdML60.3%93.3%GPT-6 Astra
MirrorCodeNot measured46.7%Only GPT-6 Astra measured
ALE-Bench1188.62951.3GPT-6 Astra
LMArena Text: coding14381489GPT-6 Astra
LMArena Text: software and IT14361474GPT-6 Astra
LMArena WebDev: React13861798GPT-6 Astra
LMArena WebDev: HTML14191775GPT-6 Astra
LMArena WebDev: from an imageNot measured1726Only GPT-6 Astra measured
SciCodeNot measured12.3%Only GPT-6 Astra measured
Neither model has a result on the other 11 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for GPT-5.4 Mini and GPT-6 Astra
BenchmarkGPT-5.4 MiniGPT-6 AstraLeader
AA Intelligence Index24.152.7GPT-6 Astra
Epoch Capabilities Index148.7168.8GPT-6 Astra
Agentic workLike-for-likePublic boards, shown as published, kept out of the index05
Agentic work benchmarks for GPT-5.4 Mini and GPT-6 Astra
BenchmarkGPT-5.4 MiniGPT-6 AstraLeader
AA Agentic Index17.951.0GPT-6 Astra
Design Arena: agents building web appsNot measured1373Only GPT-6 Astra measured
Design Arena: agents building full-stack appsNot measured1357Only GPT-6 Astra measured
Design Arena: agents building mobile appsNot measured1312Only GPT-6 Astra measured
Design Arena: agents building Godot gamesNot measured1348Only GPT-6 Astra measured
Design Arena: PowerPoint slidesNot measured1219Only GPT-6 Astra measured
Design Arena: spreadsheetsNot measured1365Only GPT-6 Astra measured
LiveBench researchNot measured85.2%Only GPT-6 Astra measured
LiveBench browserNot measured84.0%Only GPT-6 Astra measured
LiveBench long agenticNot measured82.4%Only GPT-6 Astra measured
SREBench (Vals AI)Not measured56.9%Only GPT-6 Astra measured
Finance Agent (Vals AI)45.4%53.5%GPT-6 Astra
Tax Agent Bench (Vals AI)8.8%20.7%GPT-6 Astra
Legal Research Bench (Vals AI)12.5%39.4%GPT-6 Astra
Harvey's Legal Agent Benchmark (Vals AI)Not measured7.1%Only GPT-6 Astra measured
Excel Modeling Benchmark (Vals AI)45.4%71.7%GPT-6 Astra
BioMysteryBench (Vals AI)Not measured79.3%Only GPT-6 Astra measured
Terminal-Bench-Science (Vals AI)Not measured62.9%Only GPT-6 Astra measured
Vals Science IndexNot measured65.1%Only GPT-6 Astra measured
Time Horizon Index (Vals AI)Not measured90.5%Only GPT-6 Astra measured
CUA-bench (Vals AI)Not measured19.2%Only GPT-6 Astra measured
Robo-bench (Vals AI)Not measured23.3%Only GPT-6 Astra measured
τ²-Bench Telecom83.3%Not measuredOnly GPT-5.4 Mini measured
APEX-AgentsNot measured64.7%Only GPT-6 Astra measured
Remote Labor IndexNot measured20.8%Only GPT-6 Astra measured
BALROGNot measured68.3%Only GPT-6 Astra measured
DeepResearch Bench36.3%Not measuredOnly GPT-5.4 Mini measured
SpireClimbNot measured25.7%Only GPT-6 Astra measured
Vending-Bench 2Not measured$15,515Only GPT-6 Astra measured
Neither model has a result on the other 12 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index012
Reasoning benchmarks for GPT-5.4 Mini and GPT-6 Astra
BenchmarkGPT-5.4 MiniGPT-6 AstraLeader
GPQA Diamond83.1%Not measuredOnly GPT-5.4 Mini measured
LiveBench reasoningNot measured92.7%Only GPT-6 Astra measured
MysteryMechanism (Vals AI)Not measured53.1%Only GPT-6 Astra measured
MMMU Pro (Vals AI)79.3%Not measuredOnly GPT-5.4 Mini measured
GPQA Diamond (Artificial Analysis run)87.5%96.3%GPT-6 Astra
Humanity's Last Exam (Artificial Analysis run)28.1%54.7%GPT-6 Astra
GPQA Diamond (Epoch AI run)86.9%95.8%GPT-6 Astra
ARC-AGI63.7%98.5%GPT-6 Astra
ARC-AGI-218.9%95.0%GPT-6 Astra
Chess puzzles23.0%74.0%GPT-6 Astra
Mystery game puzzles11.0%84.0%GPT-6 Astra
EBR-benchNot measured76.2%Only GPT-6 Astra measured
LMArena Text: hard prompts14241463GPT-6 Astra
LMArena Text: expert14351489GPT-6 Astra
LMArena Text: hard prompts in English14201464GPT-6 Astra
LMArena Vision: diagrams12661297GPT-6 Astra
LMArena Vision: homework12851337GPT-6 Astra
Kagi LLM Benchmark37.9%Not measuredOnly GPT-5.4 Mini measured
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index06
Math benchmarks for GPT-5.4 Mini and GPT-6 Astra
BenchmarkGPT-5.4 MiniGPT-6 AstraLeader
AIME95.6%Not measuredOnly GPT-5.4 Mini measured
OTIS Mock AIME88.9%100.0%GPT-6 Astra
FrontierMath, tiers 1 to 351.2%93.7%GPT-6 Astra
FrontierMath, tier 49.8%97.6%GPT-6 Astra
ProofBench21.0%99.0%GPT-6 Astra
LMArena Text: math14181465GPT-6 Astra
LMArena Text: mathematical work14291484GPT-6 Astra
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index05
Knowledge benchmarks for GPT-5.4 Mini and GPT-6 Astra
BenchmarkGPT-5.4 MiniGPT-6 AstraLeader
MMLU Pro84.5%Not measuredOnly GPT-5.4 Mini measured
CaseLaw (Vals AI)51.7%Not measuredOnly GPT-5.4 Mini measured
CorpFin (Vals AI)60.9%Not measuredOnly GPT-5.4 Mini measured
TaxEval (Vals AI)71.2%Not measuredOnly GPT-5.4 Mini measured
MortgageTax (Vals AI)63.5%Not measuredOnly GPT-5.4 Mini measured
MedCode (Vals AI)Not measured48.5%Only GPT-6 Astra measured
SimpleQA Verified29.4%75.6%GPT-6 Astra
LMArena Text: science14171451GPT-6 Astra
LMArena Text: medicine and healthcare13951425GPT-6 Astra
LMArena Text: legal and government14141462GPT-6 Astra
LMArena Text: business and finance14181439GPT-6 Astra
FACTS Benchmark SuiteNot measured75.7%Only GPT-6 Astra measured
Neither model has a result on the other 6 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
Web design benchmarks for GPT-5.4 Mini and GPT-6 Astra
BenchmarkGPT-5.4 MiniGPT-6 AstraLeader
Design Arena: websitesNot measured1336Only GPT-6 Astra measured
Design Arena: UI componentsNot measured1393Only GPT-6 Astra measured
Design Arena: game devNot measured1478Only GPT-6 Astra measured
Design Arena: data vizNot measured1335Only GPT-6 Astra measured
Design Arena: 3DNot measured1500Only GPT-6 Astra measured
Design Arena: all code categoriesNot measured1379Only GPT-6 Astra measured
Design Arena: SVGNot measured1431Only GPT-6 Astra measured
Design Arena: ASCII artNot measured1353Only GPT-6 Astra measured
LiveBench designNot measured82.0%Only GPT-6 Astra measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index224
General benchmarks for GPT-5.4 Mini and GPT-6 Astra
BenchmarkGPT-5.4 MiniGPT-6 AstraLeader
Vals IndexNot measured65.0%Only GPT-6 Astra measured
LiveBench data analysisNot measured83.0%Only GPT-6 Astra measured
SAGE (Vals AI)50.8%46.4%GPT-5.4 Mini
MedScribe (Vals AI)Not measured87.9%Only GPT-6 Astra measured
Vals Multimodal Index54.2%Not measuredOnly GPT-5.4 Mini measured
IFBench73.3%Not measuredOnly GPT-5.4 Mini measured
Long context reasoning (AA-LCR)77.0%80.7%GPT-6 Astra
DTBench80.0%97.3%GPT-6 Astra
LMCA40.8%64.4%GPT-6 Astra
Furniture assemblyNot measured80.0%Only GPT-6 Astra measured
LMArena Text14121441GPT-6 Astra
LMArena Text: instruction following14051455GPT-6 Astra
LMArena Text: creative writing13701420GPT-6 Astra
LMArena Vision12451280GPT-6 Astra
LMArena Text: English14121447GPT-6 Astra
LMArena Text: non-English14051432GPT-6 Astra
LMArena Text: Chinese14461489GPT-6 Astra
LMArena Text: German14081459GPT-6 Astra
LMArena Text: Japanese13751405GPT-6 Astra
LMArena Text: Korean13701428GPT-6 Astra
LMArena Text: French14411432GPT-5.4 Mini
LMArena Text: Spanish14051422GPT-6 Astra
LMArena Text: Russian14171438GPT-6 Astra
LMArena Text: Polish14151422GPT-6 Astra
LMArena Text: multi-turn14291447GPT-6 Astra
LMArena Text: longer queries14081460GPT-6 Astra
LMArena Text: writing and language13931437GPT-6 Astra
LMArena Text: entertainment and media13731427GPT-6 Astra
LMArena Vision: English12441280GPT-6 Astra
LMArena Vision: Chinese12861361GPT-6 Astra
LMArena Vision: reading text12531292GPT-6 Astra
LMArena Vision: humor1235Not measuredOnly GPT-5.4 Mini measured
LMArena Vision: recognizing things1218Not measuredOnly GPT-5.4 Mini measured
LMArena Vision: creative writing12221286GPT-6 Astra
Neither model has a result on the other 12 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 GPT-6 Astra
BenchmarkGPT-5.4 MiniGPT-6 AstraLeader
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
GPT-6 Astra$0.035Fits in one request

GPT-5.4 Mini costs less

Repository review

50K input + 3K output tokens

GPT-5.4 Mini$0.051Fits in one request
GPT-6 Astra$0.650Fits in one request

GPT-5.4 Mini costs less

Agent loop

200K input + 10K output tokens

GPT-5.4 Mini$0.195Fits in one request
GPT-6 Astra$2.50Fits in one request

GPT-5.4 Mini 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 GPT-6 Astra
SpecificationOpenAI GPT-5.4 MiniOpenAI GPT-6 Astra
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedMarch 17, 2026September 4, 2026
Context window400K1.05M
Input $/Mtok$0.75$10.00
Output $/Mtok$4.50$50.00
Coding index35.885.1
Independent measurements behind it55
Cost per solved task$0.70Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5.4 Mini better than GPT-6 Astra for coding?

Pick GPT-6 Astra for coding: it scores 85.1 on the Vibe Coding Index against 35.8 for GPT-5.4 Mini. Pick GPT-5.4 Mini for output tokens 11.1x cheaper ($4.50 vs $50.00 per million) and work measured by SAGE (Vals AI) or LMArena Text: French, where it leads.

Which is cheaper, GPT-5.4 Mini or GPT-6 Astra?

GPT-5.4 Mini costs $0.75 per million input tokens and $4.50 per million output tokens; GPT-6 Astra costs $10.00 and $50.00. Cost per solved task is known for GPT-5.4 Mini only.

Which has the bigger context window, GPT-5.4 Mini or GPT-6 Astra?

GPT-6 Astra, with 1.05M tokens against 400K.

How many benchmarks is the GPT-5.4 Mini vs GPT-6 Astra comparison based on?

71 benchmarks have a result for both models as of October 22, 2026. GPT-5.4 Mini's coding index rests on 5 independent benchmarks and GPT-6 Astra's on 5. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/gpt-5-4-mini-vs-gpt-6-astra" title="GPT-5.4 Mini vs GPT-6 Astra, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/gpt-5-4-mini-vs-gpt-6-astra">LLMBenchmark.ai, GPT-5.4 Mini vs GPT-6 Astra</a></p>