Enter opens the highlighted result. Esc closes.

GPT-5.5 vs Mistral Medium 3.5

GPT-5.5 leads Mistral Medium 3.5 on coding, 71.3 to 25.5 on the Vibe Coding Index, and Mistral Medium 3.5 costs 4.0x less per output token ($7.50 vs $30.00 per million). Updated October 21, 2026; 81 benchmarks have a result for both.

Replace GPT-5.5 Replace Mistral Medium 3.5
GPT-5.5OpenAI
Coding
71.3
Overall
69.2
Output $/M
$30.00
80
Mistral Medium 3.5Mistral AI
Coding
25.5
Overall
37.3
Output $/M
$7.50
1
vs boards won of 81 shared

GPT-5.5 or Mistral Medium 3.5: which to pick

Pick GPT-5.5 for coding: it scores 71.3 on the Vibe Coding Index against 25.5 for Mistral Medium 3.5.

Pick Mistral Medium 3.5 for output tokens 4.0x cheaper ($7.50 vs $30.00 per million) and work measured by τ²-Bench Telecom, where it leads.

Model AGPT-5.5OpenAI71.3/1005 independent benchmarks$1.65 per solved taskModel BMistral Medium 3.5Mistral AI25.5/1003 independent benchmarks$6.02 per solved task
Shared results
81
GPT-5.5 only
53
Mistral Medium 3.5 only
0
Like-for-like categories
8 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-5.5 71.3 from 5 benchmarks, Mistral Medium 3.5 25.5 from 3 benchmarks.
Widest category
Coding, 30.7 points apart on average over 13 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): GPT-5.5 69.8% against Mistral Medium 3.5 2.9%, 67.0 points apart on a 0 to 100 scale.
Output price
GPT-5.5 $30.00 against Mistral Medium 3.5 $7.50 per million output tokens, 4.0x apart.
Context window
GPT-5.5 takes 1.05M tokens, Mistral Medium 3.5 262K.

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-5.5BarsMistral Medium 3.5Ties
Coding130
Intelligence20
Agentic work51
Reasoning90
Math30
Knowledge100
Web design80
General300
All shared boards801
CodingLike-for-likeSix of these boards feed the Vibe Coding Index130
Coding benchmarks for GPT-5.5 and Mistral Medium 3.5
BenchmarkGPT-5.5Mistral Medium 3.5Leader
LiveBench coding85.1%Not measuredOnly GPT-5.5 measured
LMArena WebDev15131264GPT-5.5
AA Coding Index74.946.9GPT-5.5
LiveCodeBench (Vals AI)85.3%Not measuredOnly GPT-5.5 measured
SWE-bench Verified (Vals AI)82.6%66.4%GPT-5.5
Vibe Code Bench (Vals AI)69.8%2.9%GPT-5.5
Code migration (Vals AI)45.2%5.1%GPT-5.5
ProgramBench (Vals AI)0.5%Not measuredOnly GPT-5.5 measured
Terminal-Bench 2.1 (Vals AI)76.4%39.0%GPT-5.5
SkillsBench (Vals AI)62.5%Not measuredOnly GPT-5.5 measured
SciCode56.1%40.2%GPT-5.5
Terminal-Bench Hard (Artificial Analysis run)60.6%33.3%GPT-5.5
Terminal-Bench84.7%Not measuredOnly GPT-5.5 measured
FrontierCode43.0%Not measuredOnly GPT-5.5 measured
GSO-Bench40.2%Not measuredOnly GPT-5.5 measured
WeirdML84.9%Not measuredOnly GPT-5.5 measured
MirrorCode10.0%Not measuredOnly GPT-5.5 measured
ALE-Bench1943.0764.0GPT-5.5
LMArena Text: coding14941460GPT-5.5
LMArena Text: software and IT14931458GPT-5.5
LMArena WebDev: React15061254GPT-5.5
LMArena WebDev: HTML15431308GPT-5.5
LMArena WebDev: from an image1532Not measuredOnly GPT-5.5 measured
Neither model has a result on the other 16 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GPT-5.5 and Mistral Medium 3.5
BenchmarkGPT-5.5Mistral Medium 3.5Leader
AA Intelligence Index38.414.2GPT-5.5
Epoch Capabilities Index159.7141.4GPT-5.5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index51
Agentic work benchmarks for GPT-5.5 and Mistral Medium 3.5
BenchmarkGPT-5.5Mistral Medium 3.5Leader
AA Agentic Index36.47.0GPT-5.5
Design Arena: agents building web apps1117Not measuredOnly GPT-5.5 measured
Design Arena: agents building full-stack apps1089Not measuredOnly GPT-5.5 measured
Design Arena: agents building mobile apps1143Not measuredOnly GPT-5.5 measured
Design Arena: agents building Android apps1163Not measuredOnly GPT-5.5 measured
Design Arena: agents building Godot games1173Not measuredOnly GPT-5.5 measured
Design Arena: agents building games1179Not measuredOnly GPT-5.5 measured
Design Arena: HTML slides1067Not measuredOnly GPT-5.5 measured
Design Arena: PowerPoint slides1109Not measuredOnly GPT-5.5 measured
LiveBench research61.6%Not measuredOnly GPT-5.5 measured
LiveBench browser48.0%Not measuredOnly GPT-5.5 measured
LiveBench long agentic44.6%Not measuredOnly GPT-5.5 measured
SREBench (Vals AI)3.8%Not measuredOnly GPT-5.5 measured
Finance Agent (Vals AI)51.8%32.1%GPT-5.5
Tax Agent Bench (Vals AI)21.8%4.5%GPT-5.5
Legal Research Bench (Vals AI)40.4%9.1%GPT-5.5
Excel Modeling Benchmark (Vals AI)64.5%11.2%GPT-5.5
Time Horizon Index (Vals AI)8.3%Not measuredOnly GPT-5.5 measured
τ²-Bench Telecom93.9%94.2%Mistral Medium 3.5
APEX-Agents55.1%Not measuredOnly GPT-5.5 measured
Remote Labor Index6.3%Not measuredOnly GPT-5.5 measured
OSWorld 2.013.0%Not measuredOnly GPT-5.5 measured
DeepResearch Bench54.0%Not measuredOnly GPT-5.5 measured
PostTrainBench27.2%Not measuredOnly GPT-5.5 measured
Vending-Bench 2$7,524Not measuredOnly GPT-5.5 measured
Neither model has a result on the other 16 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index90
Reasoning benchmarks for GPT-5.5 and Mistral Medium 3.5
BenchmarkGPT-5.5Mistral Medium 3.5Leader
GPQA Diamond93.2%34.9%GPT-5.5
LiveBench reasoning89.7%Not measuredOnly GPT-5.5 measured
MMMU Pro (Vals AI)88.3%Not measuredOnly GPT-5.5 measured
GPQA Diamond (Artificial Analysis run)93.5%74.8%GPT-5.5
Humanity's Last Exam (Artificial Analysis run)45.8%13.8%GPT-5.5
GPQA Diamond (Epoch AI run)90.7%Not measuredOnly GPT-5.5 measured
ARC-AGI95.0%Not measuredOnly GPT-5.5 measured
ARC-AGI-285.0%Not measuredOnly GPT-5.5 measured
SimpleBench69.0%Not measuredOnly GPT-5.5 measured
Chess puzzles63.0%Not measuredOnly GPT-5.5 measured
Mystery game puzzles56.0%Not measuredOnly GPT-5.5 measured
EBR-bench34.3%Not measuredOnly GPT-5.5 measured
LMArena Text: hard prompts14891436GPT-5.5
LMArena Text: expert15071432GPT-5.5
LMArena Text: hard prompts in English14831445GPT-5.5
LMArena Vision: diagrams13251254GPT-5.5
LMArena Vision: homework13321230GPT-5.5
Kagi LLM Benchmark88.8%41.4%GPT-5.5
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index30
Math benchmarks for GPT-5.5 and Mistral Medium 3.5
BenchmarkGPT-5.5Mistral Medium 3.5Leader
OTIS Mock AIME84.4%Not measuredOnly GPT-5.5 measured
FrontierMath, tiers 1 to 385.3%Not measuredOnly GPT-5.5 measured
FrontierMath, tier 472.5%Not measuredOnly GPT-5.5 measured
ProofBench50.0%9.0%GPT-5.5
LMArena Text: math14871431GPT-5.5
LMArena Text: mathematical work14961440GPT-5.5
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index100
Knowledge benchmarks for GPT-5.5 and Mistral Medium 3.5
BenchmarkGPT-5.5Mistral Medium 3.5Leader
MMLU Pro88.1%75.3%GPT-5.5
LegalBench (Vals AI)86.5%Not measuredOnly GPT-5.5 measured
CaseLaw (Vals AI)66.2%44.2%GPT-5.5
CorpFin (Vals AI)68.4%58.8%GPT-5.5
TaxEval (Vals AI)75.0%68.0%GPT-5.5
MortgageTax (Vals AI)68.8%28.9%GPT-5.5
MedCode (Vals AI)49.1%33.8%GPT-5.5
SimpleQA Verified63.0%Not measuredOnly GPT-5.5 measured
LMArena Text: science14851428GPT-5.5
LMArena Text: medicine and healthcare14651409GPT-5.5
LMArena Text: legal and government14911437GPT-5.5
LMArena Text: business and finance14781426GPT-5.5
FACTS Benchmark Suite66.7%Not measuredOnly GPT-5.5 measured
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index80
Web design benchmarks for GPT-5.5 and Mistral Medium 3.5
BenchmarkGPT-5.5Mistral Medium 3.5Leader
Design Arena: websites12561142GPT-5.5
Design Arena: UI components12601079GPT-5.5
Design Arena: game dev12941079GPT-5.5
Design Arena: data viz12651120GPT-5.5
Design Arena: 3D11981061GPT-5.5
Design Arena: all code categories12601121GPT-5.5
Design Arena: SVG1234950GPT-5.5
Design Arena: ASCII art12551063GPT-5.5
LiveBench design70.3%Not measuredOnly GPT-5.5 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index300
General benchmarks for GPT-5.5 and Mistral Medium 3.5
BenchmarkGPT-5.5Mistral Medium 3.5Leader
LiveBench data analysis81.6%Not measuredOnly GPT-5.5 measured
Public Benefits Bench (Vals AI)60.9%Not measuredOnly GPT-5.5 measured
SAGE (Vals AI)51.5%37.6%GPT-5.5
MedScribe (Vals AI)86.9%67.7%GPT-5.5
Vals Multimodal Index68.1%34.8%GPT-5.5
IFBench75.9%68.8%GPT-5.5
Long context reasoning (AA-LCR)84.3%69.3%GPT-5.5
DTBench96.0%75.5%GPT-5.5
LMCA54.3%26.1%GPT-5.5
Surface Evolver Bench88.1%26.9%GPT-5.5
Furniture assembly44.2%Not measuredOnly GPT-5.5 measured
ExploitBench47.4%Not measuredOnly GPT-5.5 measured
CL-bench Life22.2%Not measuredOnly GPT-5.5 measured
LMArena Text14711421GPT-5.5
LMArena Text: instruction following14781415GPT-5.5
LMArena Text: creative writing14551374GPT-5.5
LMArena Vision12971222GPT-5.5
LMArena Text: English14701433GPT-5.5
LMArena Text: non-English14671404GPT-5.5
LMArena Text: Chinese15341444GPT-5.5
LMArena Text: German14801453GPT-5.5
LMArena Text: Japanese1499Not measuredOnly GPT-5.5 measured
LMArena Text: Korean14601384GPT-5.5
LMArena Text: French14851448GPT-5.5
LMArena Text: Spanish14681411GPT-5.5
LMArena Text: Russian14731395GPT-5.5
LMArena Text: Polish14651430GPT-5.5
LMArena Text: multi-turn14761423GPT-5.5
LMArena Text: longer queries14841415GPT-5.5
LMArena Text: writing and language14661384GPT-5.5
LMArena Text: entertainment and media14491382GPT-5.5
LMArena Vision: English12921240GPT-5.5
LMArena Vision: Chinese13601244GPT-5.5
LMArena Vision: reading text13061233GPT-5.5
LMArena Vision: humor12931216GPT-5.5
LMArena Vision: recognizing things1231Not measuredOnly GPT-5.5 measured
LMArena Vision: creative writing13011198GPT-5.5
LMArena Document1486Not measuredOnly GPT-5.5 measured
Neither model has a result on the other 8 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-5.5 and Mistral Medium 3.5
BenchmarkGPT-5.5Mistral Medium 3.5Leader
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-5.5$0.020Fits in one request
Mistral Medium 3.5$0.0053Fits in one request

Mistral Medium 3.5 costs less

Repository review

50K input + 3K output tokens

GPT-5.5$0.340Fits in one request
Mistral Medium 3.5$0.098Fits in one request

Mistral Medium 3.5 costs less

Agent loop

200K input + 10K output tokens

GPT-5.5$1.30Fits in one request
Mistral Medium 3.5$0.375Fits in one request

Mistral Medium 3.5 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-5.5 and Mistral Medium 3.5
SpecificationOpenAI GPT-5.5Mistral Medium 3.5
MakerOpenAIMistral AI
WeightsNo public weights listedNo public weights listed
ReleasedApril 24, 2026April 30, 2026
Context window1.05M262K
Input $/Mtok$5.00$1.50
Output $/Mtok$30.00$7.50
Coding index71.325.5
Independent measurements behind it53
Cost per solved task$1.65$6.02
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5.5 better than Mistral Medium 3.5 for coding?

Pick GPT-5.5 for coding: it scores 71.3 on the Vibe Coding Index against 25.5 for Mistral Medium 3.5. Pick Mistral Medium 3.5 for output tokens 4.0x cheaper ($7.50 vs $30.00 per million) and work measured by τ²-Bench Telecom, where it leads.

Which is cheaper, GPT-5.5 or Mistral Medium 3.5?

GPT-5.5 costs $5.00 per million input tokens and $30.00 per million output tokens; Mistral Medium 3.5 costs $1.50 and $7.50. Per solved SWE-bench Verified task, GPT-5.5 costs $1.65 and Mistral Medium 3.5 $6.02.

Which has the bigger context window, GPT-5.5 or Mistral Medium 3.5?

GPT-5.5, with 1.05M tokens against 262K.

How many benchmarks is the GPT-5.5 vs Mistral Medium 3.5 comparison based on?

81 benchmarks have a result for both models as of October 21, 2026. GPT-5.5's coding index rests on 5 independent benchmarks and Mistral Medium 3.5's on 3. 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-5-5-vs-mistral-medium-3-5" title="GPT-5.5 vs Mistral Medium 3.5, 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-5-5-vs-mistral-medium-3-5">LLMBenchmark.ai, GPT-5.5 vs Mistral Medium 3.5</a></p>