DeepSeek V3.2 Exp vs Mistral Medium 3.5
DeepSeek V3.2 Exp leads Mistral Medium 3.5 on the Artificial Analysis Intelligence Index, 16.6 to 14.2, and DeepSeek V3.2 Exp costs 18.3x less per output token ($0.41 vs $7.50 per million). Updated October 22, 2026; 47 benchmarks have a result for both.
DeepSeek V3.2 Exp or Mistral Medium 3.5: which to pick
Only Mistral Medium 3.5 has a coding score, 25.5 on the Vibe Coding Index; DeepSeek V3.2 Exp has none, so the two cannot be ranked on coding.
Pick DeepSeek V3.2 Exp for output tokens 18.3x cheaper ($0.41 vs $7.50 per million) and work measured by LMArena WebDev or AA Intelligence Index, where it leads.
- Shared results
- 47
- DeepSeek V3.2 Exp only
- 12
- Mistral Medium 3.5 only
- 34
- Like-for-like categories
- 8 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
Only Mistral Medium 3.5 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
DeepSeek V3.2 Exp
DeepSeek V3.2 Exp leads, 16.6 against 14.2.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Mistral Medium 3.5 has a AA Agentic Index result, 7.0.
Directional only
Reasoning
GPQA Diamond
No call
Only Mistral Medium 3.5 has a GPQA Diamond result, 34.9%.
Directional only
Knowledge
MMLU Pro
No call
Only Mistral Medium 3.5 has a MMLU Pro result, 75.3%.
Directional only
Web design
Design Arena: websites
DeepSeek V3.2 Exp
DeepSeek V3.2 Exp leads, 1178 against 1142.
Same board, both measured
Long documents
Listed context window
Mistral Medium 3.5
Mistral Medium 3.5 takes 262K tokens against 164K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V3.2 Exp
DeepSeek V3.2 Exp $0.0005, Mistral Medium 3.5 $0.0053 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Mistral Medium 3.5 has the number, $6.02.
Directional only
At a glance
- Coding index
- DeepSeek V3.2 Exp not benchmarked, Mistral Medium 3.5 25.5 from 3 benchmarks.
- Widest category
- Agentic work, 60.2 points apart on average over 1 shared benchmark.
- Widest single gap
- τ²-Bench Telecom: DeepSeek V3.2 Exp 33.9% against Mistral Medium 3.5 94.2%, 60.2 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V3.2 Exp $0.41 against Mistral Medium 3.5 $7.50 per million output tokens, 18.3x apart.
- Context window
- DeepSeek V3.2 Exp takes 164K 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.
| Category | DeepSeek V3.2 Exp | Bars | Mistral Medium 3.5 | Ties |
|---|---|---|---|---|
| Coding | 1 | 4 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 0 | 1 | ||
| Reasoning | 3 | 3 | ||
| Math | 0 | 2 | ||
| Knowledge | 2 | 2 | ||
| Web design | 7 | 0 | ||
| General | 14 | 6 | ||
| All shared boards | 29 | 18 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index14
| Benchmark | DeepSeek V3.2 Exp | Mistral Medium 3.5 | Leader |
|---|---|---|---|
| Aider polyglot | 74.2% | Not measured | Only DeepSeek V3.2 Exp measured |
| LMArena WebDev | 1272 | 1264 | DeepSeek V3.2 Exp |
| AA Coding Index | Not measured | 46.9 | Only Mistral Medium 3.5 measured |
| SWE-bench Verified (Vals AI) | Not measured | 66.4% | Only Mistral Medium 3.5 measured |
| Vibe Code Bench (Vals AI) | Not measured | 2.9% | Only Mistral Medium 3.5 measured |
| Code migration (Vals AI) | Not measured | 5.1% | Only Mistral Medium 3.5 measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 39.0% | Only Mistral Medium 3.5 measured |
| LiveCodeBench (Artificial Analysis run) | 78.9% | Not measured | Only DeepSeek V3.2 Exp measured |
| SciCode | Not measured | 40.2% | Only Mistral Medium 3.5 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 31.1% | 33.3% | Mistral Medium 3.5 |
| WeirdML | 39.5% | Not measured | Only DeepSeek V3.2 Exp measured |
| ALE-Bench | Not measured | 764.0 | Only Mistral Medium 3.5 measured |
| LMArena Text: coding | 1438 | 1460 | Mistral Medium 3.5 |
| LMArena Text: software and IT | 1438 | 1458 | Mistral Medium 3.5 |
| LMArena WebDev: React | Not measured | 1254 | Only Mistral Medium 3.5 measured |
| LMArena WebDev: HTML | 1281 | 1308 | Mistral Medium 3.5 |
| Neither model has a result on the other 23 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | DeepSeek V3.2 Exp | Mistral Medium 3.5 | Leader |
|---|---|---|---|
| AA Intelligence Index | 16.6 | 14.2 | DeepSeek V3.2 Exp |
| Epoch Capabilities Index | 144.9 | 141.4 | DeepSeek V3.2 Exp |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | DeepSeek V3.2 Exp | Mistral Medium 3.5 | Leader |
|---|---|---|---|
| AA Agentic Index | Not measured | 7.0 | Only Mistral Medium 3.5 measured |
| Finance Agent (Vals AI) | Not measured | 32.1% | Only Mistral Medium 3.5 measured |
| Tax Agent Bench (Vals AI) | Not measured | 4.5% | Only Mistral Medium 3.5 measured |
| Legal Research Bench (Vals AI) | Not measured | 9.1% | Only Mistral Medium 3.5 measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 11.2% | Only Mistral Medium 3.5 measured |
| τ²-Bench Telecom | 33.9% | 94.2% | Mistral Medium 3.5 |
| TheAgentCompany | 42.9% | Not measured | Only DeepSeek V3.2 Exp measured |
| Berkeley Function Calling Leaderboard (BFCL) | 56.7% | Not measured | Only DeepSeek V3.2 Exp measured |
| Neither model has a result on the other 33 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index33
| Benchmark | DeepSeek V3.2 Exp | Mistral Medium 3.5 | Leader |
|---|---|---|---|
| GPQA Diamond | Not measured | 34.9% | Only Mistral Medium 3.5 measured |
| GPQA Diamond (Artificial Analysis run) | 79.7% | 74.8% | DeepSeek V3.2 Exp |
| Humanity's Last Exam (Artificial Analysis run) | 14.9% | 13.8% | DeepSeek V3.2 Exp |
| LMArena Text: hard prompts | 1429 | 1436 | Mistral Medium 3.5 |
| LMArena Text: expert | 1422 | 1432 | Mistral Medium 3.5 |
| LMArena Text: hard prompts in English | 1444 | 1445 | Mistral Medium 3.5 |
| LMArena Vision: diagrams | Not measured | 1254 | Only Mistral Medium 3.5 measured |
| LMArena Vision: homework | Not measured | 1230 | Only Mistral Medium 3.5 measured |
| Kagi LLM Benchmark | 52.2% | 41.4% | DeepSeek V3.2 Exp |
| Neither model has a result on the other 13 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | DeepSeek V3.2 Exp | Mistral Medium 3.5 | Leader |
|---|---|---|---|
| AA Math Index | 87.7 | Not measured | Only DeepSeek V3.2 Exp measured |
| AIME 2025 (Artificial Analysis run) | 87.7% | Not measured | Only DeepSeek V3.2 Exp measured |
| ProofBench | Not measured | 9.0% | Only Mistral Medium 3.5 measured |
| LMArena Text: math | 1422 | 1431 | Mistral Medium 3.5 |
| LMArena Text: mathematical work | 1433 | 1440 | Mistral Medium 3.5 |
| Neither model has a result on the other 8 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index22
| Benchmark | DeepSeek V3.2 Exp | Mistral Medium 3.5 | Leader |
|---|---|---|---|
| MMLU Pro | Not measured | 75.3% | Only Mistral Medium 3.5 measured |
| CaseLaw (Vals AI) | Not measured | 44.2% | Only Mistral Medium 3.5 measured |
| CorpFin (Vals AI) | Not measured | 58.8% | Only Mistral Medium 3.5 measured |
| TaxEval (Vals AI) | Not measured | 68.0% | Only Mistral Medium 3.5 measured |
| MortgageTax (Vals AI) | Not measured | 28.9% | Only Mistral Medium 3.5 measured |
| MedCode (Vals AI) | Not measured | 33.8% | Only Mistral Medium 3.5 measured |
| MMLU Pro (Artificial Analysis run) | 85.0% | Not measured | Only DeepSeek V3.2 Exp measured |
| LMArena Text: science | 1443 | 1428 | DeepSeek V3.2 Exp |
| LMArena Text: medicine and healthcare | 1468 | 1409 | DeepSeek V3.2 Exp |
| LMArena Text: legal and government | 1419 | 1437 | Mistral Medium 3.5 |
| LMArena Text: business and finance | 1412 | 1426 | Mistral Medium 3.5 |
| Neither model has a result on the other 7 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index70
| Benchmark | DeepSeek V3.2 Exp | Mistral Medium 3.5 | Leader |
|---|---|---|---|
| Design Arena: websites | 1178 | 1142 | DeepSeek V3.2 Exp |
| Design Arena: UI components | 1164 | 1076 | DeepSeek V3.2 Exp |
| Design Arena: game dev | 1144 | 1079 | DeepSeek V3.2 Exp |
| Design Arena: data viz | 1152 | 1120 | DeepSeek V3.2 Exp |
| Design Arena: 3D | 1148 | 1060 | DeepSeek V3.2 Exp |
| Design Arena: all code categories | 1170 | 1121 | DeepSeek V3.2 Exp |
| Design Arena: SVG | 1031 | 949 | DeepSeek V3.2 Exp |
| Design Arena: ASCII art | Not measured | 1062 | Only Mistral Medium 3.5 measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index146
| Benchmark | DeepSeek V3.2 Exp | Mistral Medium 3.5 | Leader |
|---|---|---|---|
| SAGE (Vals AI) | Not measured | 37.6% | Only Mistral Medium 3.5 measured |
| MedScribe (Vals AI) | Not measured | 67.7% | Only Mistral Medium 3.5 measured |
| Vals Multimodal Index | Not measured | 34.8% | Only Mistral Medium 3.5 measured |
| IFBench | 54.1% | 68.8% | Mistral Medium 3.5 |
| Long context reasoning (AA-LCR) | 72.3% | 69.3% | DeepSeek V3.2 Exp |
| DTBench | 87.7% | 75.5% | DeepSeek V3.2 Exp |
| LMCA | 29.1% | 26.1% | DeepSeek V3.2 Exp |
| Fiction.LiveBench | 83.3% | Not measured | Only DeepSeek V3.2 Exp measured |
| CL-bench | 13.2% | Not measured | Only DeepSeek V3.2 Exp measured |
| Surface Evolver Bench | Not measured | 26.9% | Only Mistral Medium 3.5 measured |
| CL-bench Life | 9.5% | Not measured | Only DeepSeek V3.2 Exp measured |
| LMArena Text | 1425 | 1421 | DeepSeek V3.2 Exp |
| LMArena Text: instruction following | 1401 | 1415 | Mistral Medium 3.5 |
| LMArena Text: creative writing | 1404 | 1374 | DeepSeek V3.2 Exp |
| LMArena Vision | Not measured | 1222 | Only Mistral Medium 3.5 measured |
| LMArena Text: English | 1441 | 1433 | DeepSeek V3.2 Exp |
| LMArena Text: non-English | 1404 | 1404 | DeepSeek V3.2 Exp |
| LMArena Text: Chinese | 1461 | 1444 | DeepSeek V3.2 Exp |
| LMArena Text: German | 1440 | 1453 | Mistral Medium 3.5 |
| LMArena Text: Japanese | 1331 | Not measured | Only DeepSeek V3.2 Exp measured |
| LMArena Text: Korean | 1372 | 1384 | Mistral Medium 3.5 |
| LMArena Text: French | 1434 | 1448 | Mistral Medium 3.5 |
| LMArena Text: Spanish | 1441 | 1411 | DeepSeek V3.2 Exp |
| LMArena Text: Russian | 1412 | 1395 | DeepSeek V3.2 Exp |
| LMArena Text: Polish | 1435 | 1430 | DeepSeek V3.2 Exp |
| LMArena Text: multi-turn | 1422 | 1423 | Mistral Medium 3.5 |
| LMArena Text: longer queries | 1418 | 1415 | DeepSeek V3.2 Exp |
| LMArena Text: writing and language | 1407 | 1384 | DeepSeek V3.2 Exp |
| LMArena Text: entertainment and media | 1407 | 1382 | DeepSeek V3.2 Exp |
| LMArena Vision: English | Not measured | 1240 | Only Mistral Medium 3.5 measured |
| LMArena Vision: Chinese | Not measured | 1244 | Only Mistral Medium 3.5 measured |
| LMArena Vision: reading text | Not measured | 1233 | Only Mistral Medium 3.5 measured |
| LMArena Vision: humor | Not measured | 1216 | Only Mistral Medium 3.5 measured |
| LMArena Vision: creative writing | Not measured | 1198 | Only Mistral Medium 3.5 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
| Benchmark | DeepSeek V3.2 Exp | Mistral Medium 3.5 | 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
DeepSeek V3.2 Exp costs less
Repository review
50K input + 3K output tokens
DeepSeek V3.2 Exp costs less
Agent loop
200K input + 10K output tokens
DeepSeek V3.2 Exp 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 | DeepSeek V3.2 Exp | Mistral Medium 3.5 |
|---|---|---|
| Maker | DeepSeek | Mistral AI |
| Weights | Open weights | No public weights listed |
| Released | September 29, 2025 | April 30, 2026 |
| Context window | 164K | 262K |
| Input $/Mtok | $0.27 | $1.50 |
| Output $/Mtok | $0.41 | $7.50 |
| Coding index | Not benchmarked | 25.5 |
| Independent measurements behind it | 1 | 3 |
| Cost per solved task | Not computable | $6.02 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V3.2 Exp better than Mistral Medium 3.5 for coding?
Only Mistral Medium 3.5 has a coding score, 25.5 on the Vibe Coding Index; DeepSeek V3.2 Exp has none, so the two cannot be ranked on coding. Pick DeepSeek V3.2 Exp for output tokens 18.3x cheaper ($0.41 vs $7.50 per million) and work measured by LMArena WebDev or AA Intelligence Index, where it leads.
Which is cheaper, DeepSeek V3.2 Exp or Mistral Medium 3.5?
DeepSeek V3.2 Exp costs $0.27 per million input tokens and $0.41 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, DeepSeek V3.2 Exp or Mistral Medium 3.5?
Mistral Medium 3.5, with 262K tokens against 164K.
How many benchmarks is the DeepSeek V3.2 Exp vs Mistral Medium 3.5 comparison based on?
47 benchmarks have a result for both models as of October 22, 2026. DeepSeek V3.2 Exp's coding index rests on 1 independent benchmark and Mistral Medium 3.5's on 3. Where only one model has a result, the page says so instead of naming a winner.