Enter opens the highlighted result. Esc closes.

GPT-6.2 Astra vs Mistral Medium 3.5

GPT-6.2 Astra leads Mistral Medium 3.5 on coding, 87.0 to 25.4 on the Vibe Coding Index, and Mistral Medium 3.5 costs 5.3x less per output token ($7.50 vs $40.00 per million). Updated November 22, 2026; 36 benchmarks have a result for both.

GPT-6.2 AstraOpenAI
Coding
87.0
Overall
86.4
Output $/M
$40.00
26
Mistral Medium 3.5Mistral AI
Coding
25.4
Overall
33.2
Output $/M
$7.50
10
vs boards won of 36 shared

GPT-6.2 Astra or Mistral Medium 3.5: which to pick

Pick GPT-6.2 Astra for coding: it scores 87.0 on the Vibe Coding Index against 25.4 for Mistral Medium 3.5.

Pick Mistral Medium 3.5 for output tokens 5.3x cheaper ($7.50 vs $40.00 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.

Shared results
36
GPT-6.2 Astra only
40
Mistral Medium 3.5 only
51
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-6.2 Astra 87.0 from 4 benchmarks, Mistral Medium 3.5 25.4 from 3 benchmarks.
Widest category
Coding, 50.1 points apart on average over 9 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): GPT-6.2 Astra 89.9% against Mistral Medium 3.5 2.9%, 87.0 points apart on a 0 to 100 scale.
Output price
GPT-6.2 Astra $40.00 against Mistral Medium 3.5 $7.50 per million output tokens, 5.3x apart.
Context window
GPT-6.2 Astra 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-6.2 AstraBarsMistral Medium 3.5Ties
Coding72
Intelligence20
Agentic work20
Reasoning22
Knowledge13
Web design50
General73
All shared boards2610
CodingLike-for-likeSix of these boards feed the Vibe Coding Index72
Coding benchmarks for GPT-6.2 Astra and Mistral Medium 3.5
BenchmarkGPT-6.2 AstraMistral Medium 3.5Leader
LiveBench coding91.9%Not measuredOnly GPT-6.2 Astra measured
LMArena WebDev17661263GPT-6.2 Astra
SWE-bench Verified (Vals AI)Not measured66.4%Only Mistral Medium 3.5 measured
Terminal-Bench 4 (Vals AI)63.1%Not measuredOnly GPT-6.2 Astra measured
Vibe Code Bench (Vals AI)89.9%2.9%GPT-6.2 Astra
Code migration (Vals AI)Not measured5.1%Only Mistral Medium 3.5 measured
Terminal-Bench 2.1 (Vals AI)Not measured39.0%Only Mistral Medium 3.5 measured
SciCodeNot measured40.2%Only Mistral Medium 3.5 measured
Terminal-Bench Hard (Artificial Analysis run)Not measured33.3%Only Mistral Medium 3.5 measured
Terminal-Bench 4.0 (Artificial Analysis run)64.1%0.0%GPT-6.2 Astra
FrontierCode 2.049.0%1.8%GPT-6.2 Astra
MirrorCode67.8%Not measuredOnly GPT-6.2 Astra measured
ALE-BenchNot measured764.0Only Mistral Medium 3.5 measured
LMArena Text: coding14501461Mistral Medium 3.5
LMArena Text: software and IT14431458Mistral Medium 3.5
LMArena WebDev: React17701252GPT-6.2 Astra
LMArena WebDev: HTML18541310GPT-6.2 Astra
GameBench 297.227.0GPT-6.2 Astra
Neither model has a result on the other 34 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GPT-6.2 Astra and Mistral Medium 3.5
BenchmarkGPT-6.2 AstraMistral Medium 3.5Leader
AA Intelligence Index62.14.6GPT-6.2 Astra
Epoch Capabilities Index172.8141.6GPT-6.2 Astra
Agentic workLike-for-likePublic boards, shown as published, kept out of the index20
Agentic work benchmarks for GPT-6.2 Astra and Mistral Medium 3.5
BenchmarkGPT-6.2 AstraMistral Medium 3.5Leader
Design Arena: agents building web apps1334Not measuredOnly GPT-6.2 Astra measured
Design Arena: agents building full-stack apps1351Not measuredOnly GPT-6.2 Astra measured
Design Arena: agents building mobile apps1305Not measuredOnly GPT-6.2 Astra measured
Design Arena: agents building Android apps1257Not measuredOnly GPT-6.2 Astra measured
Design Arena: spreadsheets1278Not measuredOnly GPT-6.2 Astra measured
LiveBench research91.5%Not measuredOnly GPT-6.2 Astra measured
LiveBench browser93.0%Not measuredOnly GPT-6.2 Astra measured
LiveBench long agentic70.8%Not measuredOnly GPT-6.2 Astra measured
Finance Agent (Vals AI)Not measured32.1%Only Mistral Medium 3.5 measured
Tax Agent Bench (Vals AI)21.8%4.5%GPT-6.2 Astra
Legal Research Bench (Vals AI)42.8%9.1%GPT-6.2 Astra
Harvey's Legal Agent Benchmark (Vals AI)16.3%Not measuredOnly GPT-6.2 Astra measured
Excel Modeling Benchmark (Vals AI)Not measured11.2%Only Mistral Medium 3.5 measured
BioMysteryBench (Vals AI)80.7%Not measuredOnly GPT-6.2 Astra measured
Terminal-Bench-Science (Vals AI)71.4%Not measuredOnly GPT-6.2 Astra measured
Vals Science Index68.4%Not measuredOnly GPT-6.2 Astra measured
Robo-bench (Vals AI)32.7%Not measuredOnly GPT-6.2 Astra measured
τ²-Bench TelecomNot measured94.2%Only Mistral Medium 3.5 measured
APEX-Agents72.2%Not measuredOnly GPT-6.2 Astra measured
SpireClimb41.3%Not measuredOnly GPT-6.2 Astra measured
LMArena Search1191Not measuredOnly GPT-6.2 Astra measured
Neither model has a result on the other 23 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index22
Reasoning benchmarks for GPT-6.2 Astra and Mistral Medium 3.5
BenchmarkGPT-6.2 AstraMistral Medium 3.5Leader
GPQA DiamondNot measured34.9%Only Mistral Medium 3.5 measured
LiveBench reasoning91.6%Not measuredOnly GPT-6.2 Astra measured
MysteryMechanism (Vals AI)53.1%Not measuredOnly GPT-6.2 Astra measured
GPQA Diamond (Artificial Analysis run)Not measured74.8%Only Mistral Medium 3.5 measured
Humanity's Last Exam (Artificial Analysis run)Not measured13.8%Only Mistral Medium 3.5 measured
HLE-Diamond (Artificial Analysis run)72.2%4.7%GPT-6.2 Astra
CritPt v2, hard subset (Artificial Analysis run)71.5%0.5%GPT-6.2 Astra
GPQA Diamond (Epoch AI run)96.8%Not measuredOnly GPT-6.2 Astra measured
ARC-AGI98.5%Not measuredOnly GPT-6.2 Astra measured
ARC-AGI-295.8%Not measuredOnly GPT-6.2 Astra measured
SimpleBench77.3%Not measuredOnly GPT-6.2 Astra measured
Chess puzzles78.0%Not measuredOnly GPT-6.2 Astra measured
Mystery game puzzles95.0%Not measuredOnly GPT-6.2 Astra measured
EBR-bench92.4%Not measuredOnly GPT-6.2 Astra measured
LMArena Text: hard prompts14251436Mistral Medium 3.5
LMArena Text: expertNot measured1432Only Mistral Medium 3.5 measured
LMArena Text: hard prompts in English14261445Mistral Medium 3.5
LMArena Vision: diagramsNot measured1255Only Mistral Medium 3.5 measured
LMArena Vision: homeworkNot measured1229Only Mistral Medium 3.5 measured
Kagi LLM BenchmarkNot measured41.4%Only Mistral Medium 3.5 measured
Humanity's Last Exam (Scale run)66.9%Not measuredOnly GPT-6.2 Astra measured
GoBench2493Not measuredOnly GPT-6.2 Astra measured
Neither model has a result on the other 4 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Math benchmarks for GPT-6.2 Astra and Mistral Medium 3.5
BenchmarkGPT-6.2 AstraMistral Medium 3.5Leader
OTIS Mock AIME100.0%Not measuredOnly GPT-6.2 Astra measured
FrontierMath, tiers 1 to 392.6%Not measuredOnly GPT-6.2 Astra measured
FrontierMath, tier 4100.0%Not measuredOnly GPT-6.2 Astra measured
ProofBenchNot measured9.0%Only Mistral Medium 3.5 measured
LMArena Text: mathNot measured1432Only Mistral Medium 3.5 measured
LMArena Text: mathematical workNot measured1440Only Mistral Medium 3.5 measured
MathArena92.9%Not measuredOnly GPT-6.2 Astra measured
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index13
Knowledge benchmarks for GPT-6.2 Astra and Mistral Medium 3.5
BenchmarkGPT-6.2 AstraMistral Medium 3.5Leader
MMLU ProNot measured75.3%Only Mistral Medium 3.5 measured
CaseLaw (Vals AI)Not measured44.2%Only Mistral Medium 3.5 measured
CorpFin (Vals AI)Not measured58.8%Only Mistral Medium 3.5 measured
TaxEval (Vals AI)Not measured68.0%Only Mistral Medium 3.5 measured
MortgageTax (Vals AI)Not measured28.9%Only Mistral Medium 3.5 measured
MedCode (Vals AI)Not measured33.8%Only Mistral Medium 3.5 measured
AA-Omniscience50.0-38.7GPT-6.2 Astra
SimpleQA Verified78.5%Not measuredOnly GPT-6.2 Astra measured
LMArena Text: science14151427Mistral Medium 3.5
LMArena Text: medicine and healthcareNot measured1407Only Mistral Medium 3.5 measured
LMArena Text: legal and government14151437Mistral Medium 3.5
LMArena Text: business and finance14131426Mistral Medium 3.5
Neither model has a result on the other 9 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index50
Web design benchmarks for GPT-6.2 Astra and Mistral Medium 3.5
BenchmarkGPT-6.2 AstraMistral Medium 3.5Leader
Design Arena: websites13331129GPT-6.2 Astra
Design Arena: UI components13411062GPT-6.2 Astra
Design Arena: game dev15011051GPT-6.2 Astra
Design Arena: data vizNot measured1115Only Mistral Medium 3.5 measured
Design Arena: 3D14301028GPT-6.2 Astra
Design Arena: all code categories13891103GPT-6.2 Astra
Design Arena: SVGNot measured934Only Mistral Medium 3.5 measured
Design Arena: ASCII artNot measured1048Only Mistral Medium 3.5 measured
LiveBench design82.8%Not measuredOnly GPT-6.2 Astra measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index73
General benchmarks for GPT-6.2 Astra and Mistral Medium 3.5
BenchmarkGPT-6.2 AstraMistral Medium 3.5Leader
Vals Index67.7%Not measuredOnly GPT-6.2 Astra measured
LiveBench data analysis84.1%Not measuredOnly GPT-6.2 Astra measured
SAGE (Vals AI)Not measured72.0%Only Mistral Medium 3.5 measured
MedScribe (Vals AI)Not measured67.7%Only Mistral Medium 3.5 measured
Vals Multimodal IndexNot measured34.8%Only Mistral Medium 3.5 measured
IFBenchNot measured68.8%Only Mistral Medium 3.5 measured
Long context reasoning (AA-LCR)Not measured69.3%Only Mistral Medium 3.5 measured
GDPval-AANot measured769Only Mistral Medium 3.5 measured
DTBenchNot measured75.5%Only Mistral Medium 3.5 measured
LMCANot measured26.1%Only Mistral Medium 3.5 measured
Surface Evolver BenchNot measured26.9%Only Mistral Medium 3.5 measured
Furniture assembly89.2%Not measuredOnly GPT-6.2 Astra measured
LMArena Text14141421Mistral Medium 3.5
LMArena Text: instruction following14231415GPT-6.2 Astra
LMArena Text: creative writing14171373GPT-6.2 Astra
LMArena VisionNot measured1222Only Mistral Medium 3.5 measured
LMArena Text: English14171433Mistral Medium 3.5
LMArena Text: non-English14041405Mistral Medium 3.5
LMArena Text: ChineseNot measured1444Only Mistral Medium 3.5 measured
LMArena Text: GermanNot measured1460Only Mistral Medium 3.5 measured
LMArena Text: KoreanNot measured1386Only Mistral Medium 3.5 measured
LMArena Text: FrenchNot measured1449Only Mistral Medium 3.5 measured
LMArena Text: SpanishNot measured1412Only Mistral Medium 3.5 measured
LMArena Text: Russian14371397GPT-6.2 Astra
LMArena Text: PolishNot measured1429Only Mistral Medium 3.5 measured
LMArena Text: multi-turn14251423GPT-6.2 Astra
LMArena Text: longer queries14311415GPT-6.2 Astra
LMArena Text: writing and language14401383GPT-6.2 Astra
LMArena Text: entertainment and media14131382GPT-6.2 Astra
LMArena Vision: EnglishNot measured1242Only Mistral Medium 3.5 measured
LMArena Vision: ChineseNot measured1244Only Mistral Medium 3.5 measured
LMArena Vision: reading textNot measured1232Only Mistral Medium 3.5 measured
LMArena Vision: humorNot measured1218Only Mistral Medium 3.5 measured
LMArena Vision: creative writingNot measured1199Only Mistral Medium 3.5 measured
MultiChallenge73.2%Not measuredOnly GPT-6.2 Astra measured
EQ-Bench 4Not measured983Only Mistral Medium 3.5 measured
Neither model has a result on the other 13 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-6.2 Astra and Mistral Medium 3.5
BenchmarkGPT-6.2 AstraMistral 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-6.2 Astra$0.028Fits 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-6.2 Astra$0.520Fits 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-6.2 Astra$2.00Fits 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-6.2 Astra and Mistral Medium 3.5
SpecificationOpenAI GPT-6.2 AstraMistral Medium 3.5
MakerOpenAIMistral AI
WeightsNo public weights listedOpen weights
ReleasedNovember 17, 2026April 30, 2026
Context window1.05M262K
Input $/Mtok$8.00$1.50
Output $/Mtok$40.00$7.50
Coding index87.025.4
Independent measurements behind it43
Cost per solved taskNot computable$6.02
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-6.2 Astra better than Mistral Medium 3.5 for coding?

Pick GPT-6.2 Astra for coding: it scores 87.0 on the Vibe Coding Index against 25.4 for Mistral Medium 3.5. Pick Mistral Medium 3.5 for output tokens 5.3x cheaper ($7.50 vs $40.00 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.

Which is cheaper, GPT-6.2 Astra or Mistral Medium 3.5?

GPT-6.2 Astra costs $8.00 per million input tokens and $40.00 per million output tokens; Mistral Medium 3.5 costs $1.50 and $7.50. Cost per solved task is known for Mistral Medium 3.5 only.

Which has the bigger context window, GPT-6.2 Astra or Mistral Medium 3.5?

GPT-6.2 Astra, with 1.05M tokens against 262K.

How many benchmarks is the GPT-6.2 Astra vs Mistral Medium 3.5 comparison based on?

36 benchmarks have a result for both models as of November 22, 2026. GPT-6.2 Astra's coding index rests on 4 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-6-2-astra-vs-mistral-medium-3-5" title="GPT-6.2 Astra 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-6-2-astra-vs-mistral-medium-3-5">LLMBenchmark.ai, GPT-6.2 Astra vs Mistral Medium 3.5</a></p>