Skip to main content

Mistral AILLM matchup

Mistral Large 3vs.Mistral Medium 3.5

Is the higher price worth it?

Mistral Medium 3.5 performs better than Mistral Large 3, but costs 3 times (input) and 5 times (output) as much.

Mistral Medium 3.5 ahead
IntelligenceOverall index, 100 points = average
32.7

Mistral Large 3

Rank 95

46.1

Mistral Medium 3.5

Rank 89

Mistral Medium 3.5 faster
SpeedTokens per second
60

Mistral Large 3

105

Mistral Medium 3.5

Mistral Large 3 cheaper
Price$ per 1M tokens
$0.75

Mistral Large 3

$3

Mistral Medium 3.5

about as popular
Popularity0 to 100
48

Mistral Large 3

Rank 145 of 508

50

Mistral Medium 3.5

Rank 127 of 508

Choose Mistral Large 3 for

Finance (not certain)

Choose Mistral Medium 3.5 for

GeneralCodingMultimodal

Where is the difference?

Strengths at a glance

Mistral Large 3Mistral Medium 3.5Median of all ranked models

45658510525CodingMistralMedium 3.5GeneralMistralMedium 3.5MultimodalMistralMedium 3.5FinanceslightlyMistral Large3
CodingMistral Medium 3.5 ahead
Mistral Large 331.9Rank 113
Mistral Medium 3.550.1Rank 97
Median of ranked models87.5

95 % range: Mistral Large 3 19.3 to 44.5, Mistral Medium 3.5 43.3 to 56.8.

AreaMistral Large 3Mistral Medium 3.5

The same values and verdicts as the bars below. In the close-up the net starts at 25 points, just below the weaker model's lowest value; the reference group average is 100. Not in the net because a model has no own value there or the data are too thin: Reasoning, Mathematics, Knowledge, Agents, Legal, Health.

Mistral Medium 3.5 is ahead in 3 of 10 areas, Mistral Large 3 in none.

Mistral Large 3 slightly ahead: 1too little data: 6Mistral Medium 3.5 ahead: 3

Areas in the overall index
GeneralMistral Medium 3.5 +2133.4 to 54.7
CodingMistral Medium 3.5 +1831.9 to 50.1
MultimodalMistral Medium 3.5 +1332.8 to 46.1
Mathematicstoo little data22.2 to 60.4
Agentstoo little data34.3 to 54.1
Reasoningtoo little data34.3 to 28.2
Knowledgetoo little data38.4 to 36.0
Specialist fields
Financeleaning Mistral Large 3not certain · 3 and 6 tests
Healthtoo little data30.9 to 49.5
Legaltoo little data39.7 to 53.0

Bar length shows the gap in index points (full width 15 points, one standard deviation of the reference group). Solid means clearly ahead, hatched not certain yet, a thin grey bar level.

Show 1 more areas that do not count toward the overall index
Educationonly Mistral Medium 3.5 has a score

Bar length shows the gap in index points (full width 15 points, one standard deviation of the reference group). Solid means clearly ahead, hatched not certain yet, a thin grey bar level.

Which model for which task?

Everyday tasks, translated from the area values above. Each row names the values it rests on.

Writing and debugging codeMistral Medium 3.5Coding 31.9 to 50.1
Maths and data analysisopenMathematics 22.2 to 60.4 (too little data), Reasoning 34.3 to 28.2 (too little data)
Facts and researchopenKnowledge 38.4 to 36.0 (too little data)
Agents and workflowsopenAgents 34.3 to 54.1 (too little data)
Fast chat repliesMistral Medium 3.560 to 105 tokens per second
Many requests on a small budgetMistral Large 3$0.75 to $3 per 1M tokens
Finance, legal and medical textsleaning Mistral Large 3Finance 41.3 to 36.6, Legal 39.7 to 53.0 (too little data), Health 30.9 to 49.5 (too little data)
Images, screenshots and PDFsMistral Medium 3.5Multimodal 32.8 to 46.1

Mistral Large 3 vs. Mistral Medium 3.5 in our own test

SVG short test

One SVG task with the same fixed settings for every model, without hidden thinking.

Prompt

Return only one standalone SVG of a pelican riding a bicycle. Use viewBox="0 0 800 600", simple shapes, two clearly visible wheels and a recognizable pelican with a long beak. At most 30 SVG elements and 1800 characters. No Markdown, explanations, comments, scripts, animation, CSS or external resources.

Mistral Large 3

API request failed. The model was not evaluated.

Mistral Medium 3.5

Mistral Medium 3.5: SVG short test

Oct 14, 2026 · $0.0039

Test details
Test run
Oct 14, 2026
Provider
mistral
API model
mistralai/mistral-medium-3-5
Output
502 tokens
Reasoning
API reports 0 reasoning tokens
Cost of the attempt
$0.0039
Test round
43b8c5c46830
Earlier tests

Pelican on a bicycle

Prompt

Generate an SVG of a pelican riding a bicycle

Task origin: Simon Willison

Mistral Large 3

Mistral Large 3: Pelican on a bicycle
A community classic for free-form SVG drawing.

Oct 1, 2026 · 14 s · $0.0022

Test details
Run date
Oct 1, 2026
API provider
openrouter
Requested API model
mistralai/mistral-large-2512
Reasoning setting
not specified
Token limit including reasoning
8,192 tokens
Output tokens
1,451 tokens
Cost of the attempt
$0.0022
Response time
14.2 s
API model reported in response
mistralai/mistral-large-2512
Pinned endpoint
Mistral | mistralai/mistral-large-2512
Temperature
Provider default
Test protocol
community-visual-tests-v1
Technical checks
Passed, not a quality rating
Reasoning (model default)
no reasoning mode Source

Visually reviewed. This individual test does not establish overall model quality.

Earlier attempts (1)

  • Sep 7, 2026API request failedReasoning setting: not specified

Mistral Medium 3.5

Mistral Medium 3.5: Pelican on a bicycle
A community classic for free-form SVG drawing.

Oct 1, 2026 · 39 s · $0.06

Test details
Run date
Oct 1, 2026
API provider
openrouter
Requested API model
mistralai/mistral-medium-3-5
Reasoning setting
high
Token limit including reasoning
8,192 tokens
Output tokens
8,128 tokens
Cost of the attempt
$0.0610
Response time
38.7 s
API model reported in response
mistralai/mistral-medium-3-5
Pinned endpoint
Mistral | mistralai/mistral-medium-3.5-20260430
Temperature
Provider default
Test protocol
community-visual-tests-v1
Technical checks
Passed, not a quality rating

Visually reviewed. This individual test does not establish overall model quality.

Earlier attempts (1)

  • Sep 7, 2026API request failedReasoning setting: high

Mistral Large 3 and Mistral Medium 3.5 in the overall index

Every model with a place in the overall index, both models of this matchup highlighted. You can add or hide other models.

Overall index (Gradually Meta-Score v6)

20 of 95 models

Claude Fable 5.2max130.9
Gemini 4 Carbonhigh127.0
Claude Opus 5.5max125.7
Claude Fable 5.1max124.1
Claude Sonnet 5.5max121.9
Gemini 4 Argonhigh121.2
GPT-6 Astramax119.6
Muse Wonder Fullmax118.4
GPT-6.1 Solmax117.5
Claude Opus 5max117.3
Muse Wondermax113.7
Claude Fable 5max112.8
Kimi K3.1 Promax112.6
Muse Ember 1.0max110.5
GPT-5.6 Solmax109.3
GPT-6 Solmax108.4
Gemini 3.7 Flashhigh107.7
DeepSeek V4.1 Flash 1017max107.1
Mistral Medium 3.5high46.1
Mistral Large 3none32.7

Popularity compared

How often the models are searched for, used and mentioned, on a scale from 0 to 100.

Both are about as popular: Mistral Large 3 48, Mistral Medium 3.5 50 out of 100.

Over time

Weekly values from October 18, 2026 to November 16, 2026

Mistral Large 3Mistral Medium 3.5

All data side by side

The better value is bold. For the overall index only when the gap is clear.

MetricMistral Large 3Mistral Medium 3.5
Intelligence
Overall indexpoints, 100 = reference group average, higher is better32.746.1
RankRank 95 of 95Rank 89 of 95
Price per 1M tokens
Inputwhat you send$0.50$1.50
Outputwhat the model writes$1.50$7.50
Cost per task (own tests)average of 1 original-protocol task both models passed (SVG short test not included)$0.002$0.061
Speed
Tokens per second60105
Popularity
Popularitysearch, usage and mentions, 0 to 10048Rank 145 of 50850Rank 127 of 508
Technical
Context windowhow much text the model reads at once262,144 tokens≈ 390 A4 pages262,144 tokens≈ 390 A4 pages
Knowledge cutoffn/an/a
ReleasedDec 2, 2025Apr 28, 2026
AvailabilityActiveActive
Open weightsyesyes

Prices without caching or batch discounts · Page count as a rule of thumb from Artificial Analysis

Is there a better choice?

Cheaper than Mistral Large 3Muse Ember 1.0

Overall index 110.5, rank 14$0.10 input and $0.50 output per 1M tokens

View profile
Best model in the overall indexClaude Fable 5.2

Overall index 130.9, rank 1$8 input and $40 output per 1M tokens

Shared benchmarks

Single results from the same benchmark version, task and metric. A value is bold only with the same test setup and a clear gap.

A: Mistral Large 3B: Mistral Medium 3.5

CorpFin v2Overall · higher is betterTest setup not documented as identicalMistral Large 3: 61.0%Mistral Medium 3.5: 58.8%Vals AIAs of 08/12/2026
GPQA DiamondOverall · higher is betterTest setup not documented as identicalMistral Large 3: 68.4%Mistral Medium 3.5: 34.8%Vals AIAs of 09/01/2026
LMArena Code Arena WebDevOverall · higher is betterMistral Large 3: 1,231Mistral Medium 3.5: 1,263LMArenaAs of 11/15/2026
LMArena Text, Style ControlOverall · higher is betterMistral Large 3: 1,415Mistral Medium 3.5: 1,426LMArenaAs of 11/11/2026
LMArena Vision, Style ControlOverall · higher is betterMistral Large 3: 1,200Mistral Medium 3.5: 1,199LMArenaAs of 11/09/2026
MMLU ProOverall · higher is betterTest setup not documented as identicalMistral Large 3: 79.8%Mistral Medium 3.5: 75.3%Vals AIAs of 09/01/2026
MortgageTaxOverall · higher is betterTest setup not documented as identicalMistral Large 3: 52.1%Mistral Medium 3.5: 28.9%Vals AIAs of 09/01/2026
SWE-bench VerifiedOverall · higher is betterTest setup not documented as identicalMistral Large 3: 41.4%Mistral Medium 3.5: 66.4%Vals AIAs of 09/01/2026
TaxEval v2Overall · higher is betterTest setup not documented as identicalMistral Large 3: 73.1%Mistral Medium 3.5: 68.0%Vals AIAs of 09/01/2026
CorpFin v2Exact Pages · higher is betterTest setup not documented as identicalMistral Large 3: 59.8%Mistral Medium 3.5: 55.4%Vals AIAs of 08/12/2026
Show 54 more benchmarks
GPQA DiamondFew-Shot CoT · higher is betterTest setup not documented as identicalMistral Large 3: 64.1%Mistral Medium 3.5: 34.8%Vals AIAs of 09/01/2026
LMArena Code Arena WebDevHTML development · higher is betterMistral Large 3: 1,239Mistral Medium 3.5: 1,309LMArenaAs of 11/15/2026
LMArena Text, Style ControlBusiness, management and finance · higher is betterMistral Large 3: 1,411Mistral Medium 3.5: 1,431LMArenaAs of 11/11/2026
LMArena Vision, Style ControlChinese · higher is betterMistral Large 3: 1,219Mistral Medium 3.5: 1,217LMArenaAs of 11/09/2026
MMLU ProBiology · higher is betterTest setup not documented as identicalMistral Large 3: 88.6%Mistral Medium 3.5: 64.4%Vals AIAs of 09/01/2026
MortgageTaxNumerical Extraction · higher is betterTest setup not documented as identicalMistral Large 3: 50.0%Mistral Medium 3.5: 11.9%Vals AIAs of 09/01/2026
SWE-bench Verified<15 min fix · higher is betterTest setup not documented as identicalMistral Large 3: 56.2%Mistral Medium 3.5: 79.4%Vals AIAs of 09/01/2026
TaxEval v2Correctness · higher is betterTest setup not documented as identicalMistral Large 3: 61.6%Mistral Medium 3.5: 56.7%Vals AIAs of 09/01/2026
CorpFin v2Max Fitting Context · higher is betterTest setup not documented as identicalMistral Large 3: 63.8%Mistral Medium 3.5: 61.1%Vals AIAs of 08/12/2026
GPQA DiamondZero-Shot CoT · higher is betterTest setup not documented as identicalMistral Large 3: 72.7%Mistral Medium 3.5: 34.8%Vals AIAs of 09/01/2026
LMArena Code Arena WebDevWeb development · higher is betterMistral Large 3: 1,231Mistral Medium 3.5: 1,263LMArenaAs of 11/15/2026
LMArena Text, Style ControlChinese · higher is betterMistral Large 3: 1,431Mistral Medium 3.5: 1,447LMArenaAs of 11/11/2026
LMArena Vision, Style ControlCreative writing with images · higher is betterMistral Large 3: 1,198Mistral Medium 3.5: 1,196LMArenaAs of 11/09/2026
MMLU ProBusiness · higher is betterTest setup not documented as identicalMistral Large 3: 84.7%Mistral Medium 3.5: 64.4%Vals AIAs of 09/01/2026
MortgageTaxSemantic Extraction · higher is betterTest setup not documented as identicalMistral Large 3: 54.2%Mistral Medium 3.5: 45.9%Vals AIAs of 09/01/2026
SWE-bench Verified>4 hours · higher is betterTest setup not documented as identicalMistral Large 3: 33.3%Mistral Medium 3.5: 33.3%Vals AIAs of 09/01/2026
TaxEval v2Stepwise Reasoning · higher is betterTest setup not documented as identicalMistral Large 3: 84.5%Mistral Medium 3.5: 79.2%Vals AIAs of 09/01/2026
CorpFin v2Shared Max Context · higher is betterTest setup not documented as identicalMistral Large 3: 59.6%Mistral Medium 3.5: 59.9%Vals AIAs of 08/12/2026
LMArena Text, Style ControlCoding · higher is betterMistral Large 3: 1,471Mistral Medium 3.5: 1,478LMArenaAs of 11/11/2026
LMArena Vision, Style ControlDiagram understanding · higher is betterMistral Large 3: 1,233Mistral Medium 3.5: 1,242LMArenaAs of 11/09/2026
MMLU ProChemistry · higher is betterTest setup not documented as identicalMistral Large 3: 85.0%Mistral Medium 3.5: 73.1%Vals AIAs of 09/01/2026
SWE-bench Verified1-4 hours · higher is betterTest setup not documented as identicalMistral Large 3: 4.8%Mistral Medium 3.5: 33.3%Vals AIAs of 09/01/2026
LMArena Text, Style ControlCreative writing · higher is betterMistral Large 3: 1,375Mistral Medium 3.5: 1,399LMArenaAs of 11/11/2026
LMArena Vision, Style ControlEnglish · higher is betterMistral Large 3: 1,208Mistral Medium 3.5: 1,218LMArenaAs of 11/09/2026
MMLU ProComputer Science · higher is betterTest setup not documented as identicalMistral Large 3: 83.2%Mistral Medium 3.5: 82.9%Vals AIAs of 09/01/2026
SWE-bench Verified15 min - 1 hour · higher is betterTest setup not documented as identicalMistral Large 3: 36.4%Mistral Medium 3.5: 62.5%Vals AIAs of 09/01/2026
LMArena Text, Style ControlEnglish · higher is betterMistral Large 3: 1,427Mistral Medium 3.5: 1,443LMArenaAs of 11/11/2026
LMArena Vision, Style ControlHomework · higher is betterMistral Large 3: 1,203Mistral Medium 3.5: 1,216LMArenaAs of 11/09/2026
MMLU ProEconomics · higher is betterTest setup not documented as identicalMistral Large 3: 85.8%Mistral Medium 3.5: 78.2%Vals AIAs of 09/01/2026
LMArena Text, Style ControlEntertainment, sports and media · higher is betterMistral Large 3: 1,373Mistral Medium 3.5: 1,397LMArenaAs of 11/11/2026
LMArena Vision, Style ControlHumor · higher is betterMistral Large 3: 1,186Mistral Medium 3.5: 1,197LMArenaAs of 11/09/2026
MMLU ProEngineering · higher is betterTest setup not documented as identicalMistral Large 3: 74.9%Mistral Medium 3.5: 78.2%Vals AIAs of 09/01/2026
LMArena Text, Style ControlExcluding ties · higher is betterMistral Large 3: 1,402Mistral Medium 3.5: 1,420LMArenaAs of 11/11/2026
LMArena Vision, Style ControlOptical character recognition · higher is betterMistral Large 3: 1,214Mistral Medium 3.5: 1,215LMArenaAs of 11/09/2026
MMLU ProHealth · higher is betterTest setup not documented as identicalMistral Large 3: 77.3%Mistral Medium 3.5: 75.3%Vals AIAs of 09/01/2026
LMArena Text, Style ControlExpert prompts · higher is betterMistral Large 3: 1,428Mistral Medium 3.5: 1,434LMArenaAs of 11/11/2026
MMLU ProHistory · higher is betterTest setup not documented as identicalMistral Large 3: 70.9%Mistral Medium 3.5: 69.6%Vals AIAs of 09/01/2026
LMArena Text, Style ControlFrench prompts · higher is betterMistral Large 3: 1,432Mistral Medium 3.5: 1,436LMArenaAs of 11/11/2026
MMLU ProLaw · higher is betterTest setup not documented as identicalMistral Large 3: 56.4%Mistral Medium 3.5: 56.9%Vals AIAs of 09/01/2026
LMArena Text, Style ControlGerman prompts · higher is betterMistral Large 3: 1,411Mistral Medium 3.5: 1,453LMArenaAs of 11/11/2026
MMLU ProMath · higher is betterTest setup not documented as identicalMistral Large 3: 88.4%Mistral Medium 3.5: 91.0%Vals AIAs of 09/01/2026
LMArena Text, Style ControlHard prompts · higher is betterMistral Large 3: 1,433Mistral Medium 3.5: 1,445LMArenaAs of 11/11/2026
MMLU ProOthers · higher is betterTest setup not documented as identicalMistral Large 3: 79.2%Mistral Medium 3.5: 78.1%Vals AIAs of 09/01/2026
LMArena Text, Style ControlHard prompts, English · higher is betterMistral Large 3: 1,440Mistral Medium 3.5: 1,457LMArenaAs of 11/11/2026
MMLU ProPhilosophy · higher is betterTest setup not documented as identicalMistral Large 3: 76.4%Mistral Medium 3.5: 73.1%Vals AIAs of 09/01/2026
LMArena Text, Style ControlInstruction following · higher is betterMistral Large 3: 1,403Mistral Medium 3.5: 1,423LMArenaAs of 11/11/2026
MMLU ProPhysics · higher is betterTest setup not documented as identicalMistral Large 3: 86.4%Mistral Medium 3.5: 86.5%Vals AIAs of 09/01/2026
LMArena Text, Style ControlKorean prompts · higher is betterMistral Large 3: 1,364Mistral Medium 3.5: 1,388LMArenaAs of 11/11/2026
MMLU ProPsychology · higher is betterTest setup not documented as identicalMistral Large 3: 80.6%Mistral Medium 3.5: 83.0%Vals AIAs of 09/01/2026
LMArena Text, Style ControlLegal and government · higher is betterMistral Large 3: 1,422Mistral Medium 3.5: 1,440LMArenaAs of 11/11/2026
LMArena Text, Style ControlLife, physical and social science · higher is betterMistral Large 3: 1,429Mistral Medium 3.5: 1,437LMArenaAs of 11/11/2026
LMArena Text, Style ControlLonger queries · higher is betterMistral Large 3: 1,419Mistral Medium 3.5: 1,429LMArenaAs of 11/11/2026
LMArena Text, Style ControlMath · higher is betterMistral Large 3: 1,402Mistral Medium 3.5: 1,426LMArenaAs of 11/11/2026
LMArena Text, Style ControlMathematical professions · higher is betterMistral Large 3: 1,414Mistral Medium 3.5: 1,437LMArenaAs of 11/11/2026

More comparisons

Compare Mistral Large 3 and Mistral Medium 3.5 with the models people compare them with most often.

Compare another matchup