Enter opens the highlighted result. Esc closes.

Muse Spark 1.2 vs Mistral Medium 3

Muse Spark 1.2 leads Mistral Medium 3 on the Artificial Analysis Intelligence Index, 39.6 to 9.0, and Mistral Medium 3 costs 2.1x less per output token ($2.00 vs $4.25 per million). Updated October 21, 2026; 41 benchmarks have a result for both.

Replace Muse Spark 1.2 Replace Mistral Medium 3
Muse Spark 1.2Meta
Coding
55.7
Overall
60.4
Output $/M
$4.25
41
Mistral Medium 3Mistral AI
Coding
No data
Overall
No data
Output $/M
$2.00
0
vs boards won of 41 shared

Muse Spark 1.2 or Mistral Medium 3: which to pick

Only Muse Spark 1.2 has a coding score, 55.7 on the Vibe Coding Index; Mistral Medium 3 has none, so the two cannot be ranked on coding.

Pick Mistral Medium 3 for output tokens 2.1x cheaper ($2.00 vs $4.25 per million).

Model AMuse Spark 1.2Meta55.7/1005 independent benchmarks$0.64 per solved taskModel BMistral Medium 3Mistral AINot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
41
Muse Spark 1.2 only
56
Mistral Medium 3 only
16
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
Muse Spark 1.2 55.7 from 5 benchmarks, Mistral Medium 3 not benchmarked.
Widest category
Web design, 32.4 points apart on average over 6 shared benchmarks.
Widest single gap
Long context reasoning (AA-LCR): Muse Spark 1.2 79.0% against Mistral Medium 3 31.3%, 47.7 points apart on a 0 to 100 scale.
Output price
Muse Spark 1.2 $4.25 against Mistral Medium 3 $2.00 per million output tokens, 2.1x apart.
Context window
Muse Spark 1.2 takes 1.05M tokens, Mistral Medium 3 131K.

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
CategoryMuse Spark 1.2BarsMistral Medium 3Ties
Coding20
Intelligence20
Reasoning60
Math20
Knowledge40
Web design60
General190
All shared boards410
CodingLike-for-likeSix of these boards feed the Vibe Coding Index20
Coding benchmarks for Muse Spark 1.2 and Mistral Medium 3
BenchmarkMuse Spark 1.2Mistral Medium 3Leader
LiveBench coding78.2%Not measuredOnly Muse Spark 1.2 measured
LMArena WebDev1534Not measuredOnly Muse Spark 1.2 measured
AA Coding Index72.2Not measuredOnly Muse Spark 1.2 measured
SWE-bench Verified (Vals AI)86.6%Not measuredOnly Muse Spark 1.2 measured
Terminal-Bench 4 (Vals AI)6.1%Not measuredOnly Muse Spark 1.2 measured
Vibe Code Bench (Vals AI)79.1%Not measuredOnly Muse Spark 1.2 measured
Code migration (Vals AI)29.9%Not measuredOnly Muse Spark 1.2 measured
ProgramBench (Vals AI)0.5%Not measuredOnly Muse Spark 1.2 measured
IOI (Vals AI)21.8%Not measuredOnly Muse Spark 1.2 measured
Terminal-Bench 2.1 (Vals AI)69.7%Not measuredOnly Muse Spark 1.2 measured
SkillsBench (Vals AI)53.0%Not measuredOnly Muse Spark 1.2 measured
CyberBench (Vals AI)69.5%Not measuredOnly Muse Spark 1.2 measured
LiveCodeBench (Artificial Analysis run)Not measured40.0%Only Mistral Medium 3 measured
SciCode57.4%Not measuredOnly Muse Spark 1.2 measured
Terminal-Bench Hard (Artificial Analysis run)Not measured3.8%Only Mistral Medium 3 measured
FrontierSWE12.0%Not measuredOnly Muse Spark 1.2 measured
WeirdML60.3%Not measuredOnly Muse Spark 1.2 measured
LMArena Text: coding15051385Muse Spark 1.2
LMArena Text: software and IT15051385Muse Spark 1.2
LMArena WebDev: React1526Not measuredOnly Muse Spark 1.2 measured
LMArena WebDev: HTML1543Not measuredOnly Muse Spark 1.2 measured
Neither model has a result on the other 18 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Muse Spark 1.2 and Mistral Medium 3
BenchmarkMuse Spark 1.2Mistral Medium 3Leader
AA Intelligence Index39.69.0Muse Spark 1.2
Epoch Capabilities Index155.1134.0Muse Spark 1.2
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index19 boards, none shared
Agentic work benchmarks for Muse Spark 1.2 and Mistral Medium 3
BenchmarkMuse Spark 1.2Mistral Medium 3Leader
AA Agentic Index43.2Not measuredOnly Muse Spark 1.2 measured
Design Arena: agents building web apps1221Not measuredOnly Muse Spark 1.2 measured
Design Arena: agents building full-stack apps1210Not measuredOnly Muse Spark 1.2 measured
Design Arena: agents building mobile apps1181Not measuredOnly Muse Spark 1.2 measured
Design Arena: agents building Android apps1221Not measuredOnly Muse Spark 1.2 measured
Design Arena: agents building games1179Not measuredOnly Muse Spark 1.2 measured
Design Arena: HTML slides1135Not measuredOnly Muse Spark 1.2 measured
Design Arena: PowerPoint slides1198Not measuredOnly Muse Spark 1.2 measured
LiveBench research67.7%Not measuredOnly Muse Spark 1.2 measured
LiveBench browser36.0%Not measuredOnly Muse Spark 1.2 measured
LiveBench long agentic21.3%Not measuredOnly Muse Spark 1.2 measured
SREBench (Vals AI)0.0%Not measuredOnly Muse Spark 1.2 measured
Finance Agent (Vals AI)60.6%Not measuredOnly Muse Spark 1.2 measured
Tax Agent Bench (Vals AI)32.8%Not measuredOnly Muse Spark 1.2 measured
Legal Research Bench (Vals AI)43.8%Not measuredOnly Muse Spark 1.2 measured
Excel Modeling Benchmark (Vals AI)57.0%Not measuredOnly Muse Spark 1.2 measured
BioMysteryBench (Vals AI)64.8%Not measuredOnly Muse Spark 1.2 measured
τ²-Bench TelecomNot measured24.3%Only Mistral Medium 3 measured
APEX-Agents45.4%Not measuredOnly Muse Spark 1.2 measured
Neither model has a result on the other 22 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
Reasoning benchmarks for Muse Spark 1.2 and Mistral Medium 3
BenchmarkMuse Spark 1.2Mistral Medium 3Leader
LiveBench reasoning90.0%Not measuredOnly Muse Spark 1.2 measured
MMMU Pro (Vals AI)86.1%Not measuredOnly Muse Spark 1.2 measured
GPQA Diamond (Artificial Analysis run)90.4%57.8%Muse Spark 1.2
Humanity's Last Exam (Artificial Analysis run)45.5%4.1%Muse Spark 1.2
SimpleBench74.5%Not measuredOnly Muse Spark 1.2 measured
Chess puzzles27.0%Not measuredOnly Muse Spark 1.2 measured
LMArena Text: hard prompts14891364Muse Spark 1.2
LMArena Text: expert14751342Muse Spark 1.2
LMArena Text: hard prompts in English14871374Muse Spark 1.2
LMArena Vision: diagrams13171155Muse Spark 1.2
LMArena Vision: homeworkNot measured1165Only Mistral Medium 3 measured
Neither model has a result on the other 11 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index20
Math benchmarks for Muse Spark 1.2 and Mistral Medium 3
BenchmarkMuse Spark 1.2Mistral Medium 3Leader
AA Math IndexNot measured30.3Only Mistral Medium 3 measured
AIME 2025 (Artificial Analysis run)Not measured30.3%Only Mistral Medium 3 measured
ProofBench43.0%Not measuredOnly Muse Spark 1.2 measured
LMArena Text: math14841353Muse Spark 1.2
LMArena Text: mathematical work14801354Muse Spark 1.2
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
Knowledge benchmarks for Muse Spark 1.2 and Mistral Medium 3
BenchmarkMuse Spark 1.2Mistral Medium 3Leader
MMLU Pro88.3%Not measuredOnly Muse Spark 1.2 measured
LegalBench (Vals AI)85.3%Not measuredOnly Muse Spark 1.2 measured
CorpFin (Vals AI)70.9%Not measuredOnly Muse Spark 1.2 measured
TaxEval (Vals AI)80.4%Not measuredOnly Muse Spark 1.2 measured
MortgageTax (Vals AI)65.4%Not measuredOnly Muse Spark 1.2 measured
MedCode (Vals AI)49.4%Not measuredOnly Muse Spark 1.2 measured
MMLU Pro (Artificial Analysis run)Not measured76.0%Only Mistral Medium 3 measured
SimpleQA Verified60.3%Not measuredOnly Muse Spark 1.2 measured
LMArena Text: science15001381Muse Spark 1.2
LMArena Text: medicine and healthcare14961381Muse Spark 1.2
LMArena Text: legal and government15171384Muse Spark 1.2
LMArena Text: business and finance14841352Muse Spark 1.2
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index60
Web design benchmarks for Muse Spark 1.2 and Mistral Medium 3
BenchmarkMuse Spark 1.2Mistral Medium 3Leader
Design Arena: websites13131079Muse Spark 1.2
Design Arena: UI components13011024Muse Spark 1.2
Design Arena: game dev13011018Muse Spark 1.2
Design Arena: data viz13341034Muse Spark 1.2
Design Arena: 3D13021084Muse Spark 1.2
Design Arena: all code categories13131069Muse Spark 1.2
Design Arena: SVG1286Not measuredOnly Muse Spark 1.2 measured
Design Arena: ASCII art1283Not measuredOnly Muse Spark 1.2 measured
LiveBench design67.1%Not measuredOnly Muse Spark 1.2 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index190
General benchmarks for Muse Spark 1.2 and Mistral Medium 3
BenchmarkMuse Spark 1.2Mistral Medium 3Leader
LiveBench data analysis76.5%Not measuredOnly Muse Spark 1.2 measured
Public Benefits Bench (Vals AI)68.5%Not measuredOnly Muse Spark 1.2 measured
SAGE (Vals AI)47.7%Not measuredOnly Muse Spark 1.2 measured
MedScribe (Vals AI)90.1%Not measuredOnly Muse Spark 1.2 measured
Vals Multimodal Index69.8%Not measuredOnly Muse Spark 1.2 measured
IFBenchNot measured39.3%Only Mistral Medium 3 measured
Long context reasoning (AA-LCR)79.0%31.3%Muse Spark 1.2
DTBench94.7%62.3%Muse Spark 1.2
LMCA48.4%Not measuredOnly Muse Spark 1.2 measured
LMArena Text14821370Muse Spark 1.2
LMArena Text: instruction following14621338Muse Spark 1.2
LMArena Text: creative writing14361343Muse Spark 1.2
LMArena Vision13041157Muse Spark 1.2
LMArena Text: English14831385Muse Spark 1.2
LMArena Text: non-English14761353Muse Spark 1.2
LMArena Text: Chinese15091374Muse Spark 1.2
LMArena Text: GermanNot measured1377Only Mistral Medium 3 measured
LMArena Text: JapaneseNot measured1316Only Mistral Medium 3 measured
LMArena Text: Korean14621304Muse Spark 1.2
LMArena Text: French14951368Muse Spark 1.2
LMArena Text: Spanish14841364Muse Spark 1.2
LMArena Text: Russian14831359Muse Spark 1.2
LMArena Text: PolishNot measured1371Only Mistral Medium 3 measured
LMArena Text: multi-turn15021384Muse Spark 1.2
LMArena Text: longer queries14731360Muse Spark 1.2
LMArena Text: writing and language14531352Muse Spark 1.2
LMArena Text: entertainment and media14431335Muse Spark 1.2
LMArena Vision: English12911172Muse Spark 1.2
LMArena Vision: ChineseNot measured1161Only Mistral Medium 3 measured
LMArena Vision: reading text13131154Muse Spark 1.2
LMArena Vision: humorNot measured1145Only Mistral Medium 3 measured
LMArena Vision: captioningNot measured1094Only Mistral Medium 3 measured
LMArena Vision: recognizing thingsNot measured1141Only Mistral Medium 3 measured
LMArena Vision: creative writingNot measured1146Only Mistral Medium 3 measured
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 Muse Spark 1.2 and Mistral Medium 3
BenchmarkMuse Spark 1.2Mistral Medium 3Leader
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

Muse Spark 1.2$0.0034Fits in one request
Mistral Medium 3$0.0014Fits in one request

Mistral Medium 3 costs less

Repository review

50K input + 3K output tokens

Muse Spark 1.2$0.075Fits in one request
Mistral Medium 3$0.026Fits in one request

Mistral Medium 3 costs less

Agent loop

200K input + 10K output tokens

Muse Spark 1.2$0.292Fits in one request
Mistral Medium 3$0.100Exceeds the context window

Mistral Medium 3 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 Muse Spark 1.2 and Mistral Medium 3
SpecificationMeta Muse Spark 1.2Mistral Medium 3
MakerMetaMistral AI
WeightsNo public weights listedNo public weights listed
ReleasedAugust 5, 2026May 7, 2025
Context window1.05M131K
Input $/Mtok$1.25$0.40
Output $/Mtok$4.25$2.00
Coding index55.7Not benchmarked
Independent measurements behind it50
Cost per solved task$0.64Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Muse Spark 1.2 better than Mistral Medium 3 for coding?

Only Muse Spark 1.2 has a coding score, 55.7 on the Vibe Coding Index; Mistral Medium 3 has none, so the two cannot be ranked on coding. Pick Mistral Medium 3 for output tokens 2.1x cheaper ($2.00 vs $4.25 per million).

Which is cheaper, Muse Spark 1.2 or Mistral Medium 3?

Muse Spark 1.2 costs $1.25 per million input tokens and $4.25 per million output tokens; Mistral Medium 3 costs $0.40 and $2.00. Cost per solved task is known for Muse Spark 1.2 only.

Which has the bigger context window, Muse Spark 1.2 or Mistral Medium 3?

Muse Spark 1.2, with 1.05M tokens against 131K.

How many benchmarks is the Muse Spark 1.2 vs Mistral Medium 3 comparison based on?

41 benchmarks have a result for both models as of October 21, 2026. Muse Spark 1.2's coding index rests on 5 independent benchmarks and Mistral Medium 3's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next