Mistral Medium 3 vs Qwen3 30B A3B Instruct 2507
Mistral Medium 3 leads Qwen3 30B A3B Instruct 2507 on the Artificial Analysis Intelligence Index, 9.0 to 7.5, and Qwen3 30B A3B Instruct 2507 costs 6.7x less per output token ($0.30 vs $2.00 per million). Updated October 22, 2026; 40 benchmarks have a result for both.
Mistral Medium 3 or Qwen3 30B A3B Instruct 2507: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Mistral Medium 3 for work measured by LiveCodeBench (Artificial Analysis run) or Terminal-Bench Hard (Artificial Analysis run), where it leads.
- Shared results
- 40
- Mistral Medium 3 only
- 17
- Qwen3 30B A3B Instruct 2507 only
- 5
- 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.
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 7.5.
Same board, both measured
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
Qwen3 30B A3B Instruct 2507
Qwen3 30B A3B Instruct 2507 takes 262K tokens against 131K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3 30B A3B Instruct 2507
Mistral Medium 3 $0.0014, Qwen3 30B A3B Instruct 2507 $0.0002 per turn.
List prices
At a glance
- Coding index
- Mistral Medium 3 not benchmarked, Qwen3 30B A3B Instruct 2507 not benchmarked.
- Widest category
- Math, 22.1 points apart on average over 4 shared benchmarks.
- Widest single gap
- AA Math Index: Mistral Medium 3 30.3 against Qwen3 30B A3B Instruct 2507 66.3, 36.0 points apart on a 0 to 100 scale.
- Output price
- Mistral Medium 3 $2.00 against Qwen3 30B A3B Instruct 2507 $0.30 per million output tokens, 6.7x apart.
- Context window
- Mistral Medium 3 takes 131K tokens, Qwen3 30B A3B Instruct 2507 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.
| Category | Mistral Medium 3 | Bars | Qwen3 30B A3B Instruct 2507 | Ties |
|---|---|---|---|---|
| Coding | 0 | 4 | ||
| Intelligence | 1 | 0 | ||
| Agentic work | 0 | 1 | ||
| Reasoning | 0 | 5 | ||
| Math | 0 | 4 | ||
| Knowledge | 0 | 5 | ||
| General | 5 | 15 | ||
| All shared boards | 6 | 34 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index04
| Benchmark | Mistral Medium 3 | Qwen3 30B A3B Instruct 2507 | Leader |
|---|---|---|---|
| LiveCodeBench (Artificial Analysis run) | 40.0% | 70.7% | Qwen3 30B A3B Instruct 2507 |
| SciCode | Not measured | 33.0% | Only Qwen3 30B A3B Instruct 2507 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 3.8% | 6.1% | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: coding | 1385 | 1416 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: software and IT | 1385 | 1417 | Qwen3 30B A3B Instruct 2507 |
| Neither model has a result on the other 34 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Mistral Medium 3 | Qwen3 30B A3B Instruct 2507 | Leader |
|---|---|---|---|
| AA Intelligence Index | 9.0 | 7.5 | Mistral Medium 3 |
| Epoch Capabilities Index | 134.0 | Not measured | Only Mistral Medium 3 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Mistral Medium 3 | Qwen3 30B A3B Instruct 2507 | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 24.3% | 28.1% | Qwen3 30B A3B Instruct 2507 |
| Berkeley Function Calling Leaderboard (BFCL) | Not measured | 41.4% | Only Qwen3 30B A3B Instruct 2507 measured |
| Neither model has a result on the other 39 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Mistral Medium 3 | Qwen3 30B A3B Instruct 2507 | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 57.8% | 70.7% | Qwen3 30B A3B Instruct 2507 |
| Humanity's Last Exam (Artificial Analysis run) | 4.1% | 10.3% | Qwen3 30B A3B Instruct 2507 |
| GPQA Diamond (Epoch AI run) | Not measured | 55.6% | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: hard prompts | 1364 | 1398 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: expert | 1342 | 1397 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: hard prompts in English | 1374 | 1406 | Qwen3 30B A3B Instruct 2507 |
| LMArena Vision: diagrams | 1155 | Not measured | Only Mistral Medium 3 measured |
| LMArena Vision: homework | 1165 | Not measured | Only Mistral Medium 3 measured |
| Neither model has a result on the other 14 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index04
| Benchmark | Mistral Medium 3 | Qwen3 30B A3B Instruct 2507 | Leader |
|---|---|---|---|
| AA Math Index | 30.3 | 66.3 | Qwen3 30B A3B Instruct 2507 |
| AIME 2025 (Artificial Analysis run) | 30.3% | 66.3% | Qwen3 30B A3B Instruct 2507 |
| OTIS Mock AIME | Not measured | 62.2% | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: math | 1353 | 1394 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: mathematical work | 1354 | 1411 | Qwen3 30B A3B Instruct 2507 |
| Neither model has a result on the other 8 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Mistral Medium 3 | Qwen3 30B A3B Instruct 2507 | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | 76.0% | 80.5% | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: science | 1381 | 1393 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: medicine and healthcare | 1381 | 1392 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: legal and government | 1384 | 1387 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: business and finance | 1352 | 1389 | Qwen3 30B A3B Instruct 2507 |
| Neither model has a result on the other 13 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Mistral Medium 3 | Qwen3 30B A3B Instruct 2507 | Leader |
|---|---|---|---|
| Design Arena: websites | 1079 | Not measured | Only Mistral Medium 3 measured |
| Design Arena: UI components | 1022 | Not measured | Only Mistral Medium 3 measured |
| Design Arena: game dev | 1017 | Not measured | Only Mistral Medium 3 measured |
| Design Arena: data viz | 1034 | Not measured | Only Mistral Medium 3 measured |
| Design Arena: 3D | 1083 | Not measured | Only Mistral Medium 3 measured |
| Design Arena: all code categories | 1068 | Not measured | 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 index515
| Benchmark | Mistral Medium 3 | Qwen3 30B A3B Instruct 2507 | Leader |
|---|---|---|---|
| IFBench | 39.3% | 50.7% | Qwen3 30B A3B Instruct 2507 |
| Long context reasoning (AA-LCR) | 31.3% | 61.3% | Qwen3 30B A3B Instruct 2507 |
| DTBench | 62.3% | 67.2% | Qwen3 30B A3B Instruct 2507 |
| LMCA | Not measured | 22.4% | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text | 1370 | 1384 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: instruction following | 1338 | 1363 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: creative writing | 1343 | 1317 | Mistral Medium 3 |
| LMArena Vision | 1157 | Not measured | Only Mistral Medium 3 measured |
| LMArena Text: English | 1385 | 1391 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: non-English | 1353 | 1372 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: Chinese | 1374 | 1432 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: German | 1377 | 1379 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: Japanese | 1316 | 1337 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: Korean | 1304 | 1330 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: French | 1368 | 1419 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: Spanish | 1364 | 1404 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: Russian | 1359 | 1369 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: Polish | 1371 | 1353 | Mistral Medium 3 |
| LMArena Text: multi-turn | 1384 | 1378 | Mistral Medium 3 |
| LMArena Text: longer queries | 1360 | 1378 | Qwen3 30B A3B Instruct 2507 |
| LMArena Text: writing and language | 1352 | 1336 | Mistral Medium 3 |
| LMArena Text: entertainment and media | 1335 | 1334 | Mistral Medium 3 |
| LMArena Vision: English | 1172 | Not measured | Only Mistral Medium 3 measured |
| LMArena Vision: Chinese | 1161 | Not measured | Only Mistral Medium 3 measured |
| LMArena Vision: reading text | 1154 | Not measured | Only Mistral Medium 3 measured |
| LMArena Vision: humor | 1145 | Not measured | Only Mistral Medium 3 measured |
| LMArena Vision: captioning | 1094 | Not measured | Only Mistral Medium 3 measured |
| LMArena Vision: recognizing things | 1141 | Not measured | Only Mistral Medium 3 measured |
| LMArena Vision: creative writing | 1146 | Not measured | Only Mistral Medium 3 measured |
| Neither model has a result on the other 17 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Mistral Medium 3 | Qwen3 30B A3B Instruct 2507 | 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
Qwen3 30B A3B Instruct 2507 costs less
Repository review
50K input + 3K output tokens
Qwen3 30B A3B Instruct 2507 costs less
Agent loop
200K input + 10K output tokens
Qwen3 30B A3B Instruct 2507 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 Medium 3 | Qwen3 30B A3B Instruct 2507 |
|---|---|---|
| Maker | Mistral AI | Qwen |
| Weights | No public weights listed | Open weights |
| Released | May 7, 2025 | July 29, 2025 |
| Context window | 131K | 262K |
| Input $/Mtok | $0.40 | $0.09 |
| Output $/Mtok | $2.00 | $0.30 |
| 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 Medium 3 better than Qwen3 30B A3B Instruct 2507 for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Mistral Medium 3 for work measured by LiveCodeBench (Artificial Analysis run) or Terminal-Bench Hard (Artificial Analysis run), where it leads.
Which is cheaper, Mistral Medium 3 or Qwen3 30B A3B Instruct 2507?
Mistral Medium 3 costs $0.40 per million input tokens and $2.00 per million output tokens; Qwen3 30B A3B Instruct 2507 costs $0.09 and $0.30. Cost per solved task is known for neither of them.
Which has the bigger context window, Mistral Medium 3 or Qwen3 30B A3B Instruct 2507?
Qwen3 30B A3B Instruct 2507, with 262K tokens against 131K.
How many benchmarks is the Mistral Medium 3 vs Qwen3 30B A3B Instruct 2507 comparison based on?
40 benchmarks have a result for both models as of October 22, 2026. Mistral Medium 3's coding index rests on 0 independent benchmarks and Qwen3 30B A3B Instruct 2507's on 0. Where only one model has a result, the page says so instead of naming a winner.