Mistral Large vs Mistral Medium 3
Updated October 21, 2026. 35 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Neither model has a coding score yet, so the two cannot be ranked on coding.
Mistral Large has the edge where what matters is LiveCodeBench (Artificial Analysis run) or LMArena Text: coding, where it leads.
- Shared results
- 35
- Mistral Large only
- 9
- Mistral Medium 3 only
- 22
- Like-for-like categories
- 6 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Neither has a coding score yet.
One side not measured
Intelligence
AA Intelligence Index
Mistral Medium 3
Mistral Medium 3 leads, 9.0 against 5.8.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only Mistral Large has a GPQA Diamond result, 47.7%.
Directional only
Math
AIME
No call
Only Mistral Large has a AIME result, 9.2%.
Directional only
Knowledge
MMLU Pro
No call
Only Mistral Large has a MMLU Pro result, 69.7%.
Directional only
Web design
Design Arena: websites
No call
Only Mistral Medium 3 has a Design Arena: websites result, 1079.
Directional only
Long documents
Listed context window
Mistral Medium 3
Mistral Medium 3 takes 131K tokens against 128K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Mistral Medium 3
Mistral Large $0.0050, Mistral Medium 3 $0.0014 per turn.
List prices
At a glance
- Coding index
- Mistral Large not benchmarked, Mistral Medium 3 not benchmarked.
- Widest category
- Knowledge, 20.6 points apart on average over 5 shared benchmarks.
- Widest single gap
- LMArena Text: Polish: Mistral Large 1157 against Mistral Medium 3 1371, 35.6 points apart on a 0 to 100 scale.
- Output price
- Mistral Large $6.00 against Mistral Medium 3 $2.00 per million output tokens, 3.0x apart.
- Context window
- Mistral Large takes 128K 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.
| Category | Mistral Large | Bars | Mistral Medium 3 | Ties |
|---|---|---|---|---|
| Coding | 0 | 3 | ||
| Intelligence | 0 | 2 | ||
| Reasoning | 0 | 5 | ||
| Math | 0 | 2 | ||
| Knowledge | 0 | 5 | ||
| General | 0 | 18 | ||
| All shared boards | 0 | 35 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index03
| Benchmark | Mistral Large | Mistral Medium 3 | Leader |
|---|---|---|---|
| LiveCodeBench (Vals AI) | 37.1% | Not measured | Only Mistral Large measured |
| LiveCodeBench (Artificial Analysis run) | 17.8% | 40.0% | Mistral Medium 3 |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 3.8% | Only Mistral Medium 3 measured |
| LMArena Text: coding | 1275 | 1385 | Mistral Medium 3 |
| LMArena Text: software and IT | 1274 | 1385 | Mistral Medium 3 |
| Neither model has a result on the other 33 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Mistral Large | Mistral Medium 3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 5.8 | 9.0 | Mistral Medium 3 |
| Epoch Capabilities Index | 122.1 | 134.0 | Mistral Medium 3 |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Mistral Large | Mistral Medium 3 | Leader |
|---|---|---|---|
| τ²-Bench Telecom | Not measured | 24.3% | Only Mistral Medium 3 measured |
| Neither model has a result on the other 38 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Mistral Large | Mistral Medium 3 | Leader |
|---|---|---|---|
| GPQA Diamond | 47.7% | Not measured | Only Mistral Large measured |
| GPQA Diamond (Artificial Analysis run) | 35.1% | 57.8% | Mistral Medium 3 |
| Humanity's Last Exam (Artificial Analysis run) | 3.5% | 4.1% | Mistral Medium 3 |
| LMArena Text: hard prompts | 1257 | 1364 | Mistral Medium 3 |
| LMArena Text: expert | 1208 | 1342 | Mistral Medium 3 |
| LMArena Text: hard prompts in English | 1268 | 1374 | Mistral Medium 3 |
| LMArena Vision: diagrams | Not measured | 1155 | Only Mistral Medium 3 measured |
| LMArena Vision: homework | Not measured | 1165 | Only Mistral Medium 3 measured |
| Neither model has a result on the other 13 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Mistral Large | Mistral Medium 3 | Leader |
|---|---|---|---|
| AIME | 9.2% | Not measured | Only Mistral Large measured |
| MATH 500 (Vals AI) | 74.4% | Not measured | Only Mistral Large measured |
| MGSM (Vals AI) | 87.2% | Not measured | Only Mistral Large measured |
| AA Math Index | Not measured | 30.3 | Only Mistral Medium 3 measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 30.3% | Only Mistral Medium 3 measured |
| LMArena Text: math | 1261 | 1353 | Mistral Medium 3 |
| LMArena Text: mathematical work | 1257 | 1354 | Mistral Medium 3 |
| Neither model has a result on the other 6 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Mistral Large | Mistral Medium 3 | Leader |
|---|---|---|---|
| MMLU Pro | 69.7% | Not measured | Only Mistral Large measured |
| TaxEval (Vals AI) | 63.8% | Not measured | Only Mistral Large measured |
| MedQA (Vals AI) | 76.2% | Not measured | Only Mistral Large measured |
| MMLU Pro (Artificial Analysis run) | 51.5% | 76.0% | Mistral Medium 3 |
| MMLU | 68.8% | Not measured | Only Mistral Large measured |
| LMArena Text: science | 1260 | 1381 | Mistral Medium 3 |
| LMArena Text: medicine and healthcare | 1255 | 1381 | Mistral Medium 3 |
| LMArena Text: legal and government | 1275 | 1384 | Mistral Medium 3 |
| LMArena Text: business and finance | 1238 | 1352 | Mistral Medium 3 |
| Neither model has a result on the other 8 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Mistral Large | Mistral Medium 3 | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1079 | Only Mistral Medium 3 measured |
| Design Arena: UI components | Not measured | 1024 | Only Mistral Medium 3 measured |
| Design Arena: game dev | Not measured | 1018 | Only Mistral Medium 3 measured |
| Design Arena: data viz | Not measured | 1034 | Only Mistral Medium 3 measured |
| Design Arena: 3D | Not measured | 1083 | Only Mistral Medium 3 measured |
| Design Arena: all code categories | Not measured | 1069 | Only Mistral Medium 3 measured |
| Neither model has a result on the other 3 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index018
| Benchmark | Mistral Large | Mistral Medium 3 | Leader |
|---|---|---|---|
| IFBench | Not measured | 39.3% | Only Mistral Medium 3 measured |
| Long context reasoning (AA-LCR) | Not measured | 31.3% | Only Mistral Medium 3 measured |
| DTBench | 60.8% | 62.3% | Mistral Medium 3 |
| LMArena Text | 1265 | 1370 | Mistral Medium 3 |
| LMArena Text: instruction following | 1249 | 1338 | Mistral Medium 3 |
| LMArena Text: creative writing | 1242 | 1343 | Mistral Medium 3 |
| LMArena Vision | Not measured | 1157 | Only Mistral Medium 3 measured |
| LMArena Text: English | 1284 | 1385 | Mistral Medium 3 |
| LMArena Text: non-English | 1235 | 1353 | Mistral Medium 3 |
| LMArena Text: Chinese | 1239 | 1374 | Mistral Medium 3 |
| LMArena Text: German | 1242 | 1377 | Mistral Medium 3 |
| LMArena Text: Japanese | 1170 | 1316 | Mistral Medium 3 |
| LMArena Text: Korean | 1202 | 1304 | Mistral Medium 3 |
| LMArena Text: French | 1325 | 1368 | Mistral Medium 3 |
| LMArena Text: Spanish | 1268 | 1364 | Mistral Medium 3 |
| LMArena Text: Russian | 1252 | 1359 | Mistral Medium 3 |
| LMArena Text: Polish | 1157 | 1371 | Mistral Medium 3 |
| LMArena Text: multi-turn | 1260 | 1384 | Mistral Medium 3 |
| LMArena Text: longer queries | 1261 | 1360 | Mistral Medium 3 |
| LMArena Text: writing and language | 1261 | 1352 | Mistral Medium 3 |
| LMArena Text: entertainment and media | 1230 | 1335 | Mistral Medium 3 |
| LMArena Vision: English | Not measured | 1172 | Only Mistral Medium 3 measured |
| LMArena Vision: Chinese | Not measured | 1161 | Only Mistral Medium 3 measured |
| LMArena Vision: reading text | Not measured | 1154 | Only Mistral Medium 3 measured |
| LMArena Vision: humor | Not measured | 1145 | Only Mistral Medium 3 measured |
| LMArena Vision: captioning | Not measured | 1094 | Only Mistral Medium 3 measured |
| LMArena Vision: recognizing things | Not measured | 1141 | Only Mistral Medium 3 measured |
| LMArena Vision: creative writing | Not measured | 1146 | Only Mistral Medium 3 measured |
| Neither model has a result on the other 18 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Mistral Large | Mistral Medium 3 | Leader |
|---|---|---|---|
| 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
Mistral Medium 3 costs less
Repository review
50K input + 3K output tokens
Mistral Medium 3 costs less
Agent loop
200K input + 10K output tokens
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.
| Specification | Mistral Large | Mistral Medium 3 |
|---|---|---|
| Maker | Mistral AI | Mistral AI |
| Weights | No public weights listed | No public weights listed |
| Released | February 26, 2024 | May 7, 2025 |
| Context window | 128K | 131K |
| Input $/Mtok | $2.00 | $0.40 |
| Output $/Mtok | $6.00 | $2.00 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 0 | 0 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Mistral Large better than Mistral Medium 3 for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Mistral Large has the edge where what matters is LiveCodeBench (Artificial Analysis run) or LMArena Text: coding, where it leads.
Which is cheaper, Mistral Large or Mistral Medium 3?
Mistral Large costs $2.00 per million input tokens and $6.00 per million output tokens; Mistral Medium 3 costs $0.40 and $2.00. Cost per solved task is known for neither of them.
Which has the bigger context window, Mistral Large or Mistral Medium 3?
Mistral Medium 3, with 131K tokens against 128K.
How many benchmarks is the Mistral Large vs Mistral Medium 3 comparison based on?
35 benchmarks have a result for both models as of October 21, 2026. Mistral Large's coding index rests on 0 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.