Mistral Medium 3.5 vs Hy3 preview
Hy3 preview leads Mistral Medium 3.5 on the Artificial Analysis Intelligence Index, 25.3 to 14.2, and Hy3 preview costs 12.5x less per output token ($0.60 vs $7.50 per million). Updated October 21, 2026; 36 benchmarks have a result for both.
Mistral Medium 3.5 or Hy3 preview: which to pick
Only Mistral Medium 3.5 has a coding score, 25.5 on the Vibe Coding Index; Hy3 preview has none, so the two cannot be ranked on coding.
Pick Hy3 preview for output tokens 12.5x cheaper ($0.60 vs $7.50 per million) and work measured by LMArena WebDev or AA Coding Index, where it leads.
- Shared results
- 36
- Mistral Medium 3.5 only
- 45
- Hy3 preview only
- 0
- 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
Only Mistral Medium 3.5 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
Hy3 preview
Hy3 preview leads, 25.3 against 14.2.
Same board, both measured
Agentic work
AA Agentic Index
Hy3 preview
Hy3 preview leads, 24.1 against 7.0.
Same board, both measured
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
No call
Only Mistral Medium 3.5 has a Design Arena: websites result, 1142.
Directional only
Long documents
Listed context window
No call
Both take 262K tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Hy3 preview
Mistral Medium 3.5 $0.0053, Hy3 preview $0.0005 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
- Mistral Medium 3.5 25.5 from 3 benchmarks, Hy3 preview not benchmarked.
- Widest category
- Intelligence, 11.1 points apart on average over 1 shared benchmark.
- Widest single gap
- AA Agentic Index: Mistral Medium 3.5 7.0 against Hy3 preview 24.1, 17.1 points apart on a 0 to 100 scale.
- Output price
- Mistral Medium 3.5 $7.50 against Hy3 preview $0.60 per million output tokens, 12.5x apart.
- Context window
- Both take 262K tokens.
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 3.5 | Bars | Hy3 preview | Ties |
|---|---|---|---|---|
| Coding | 2 | 5 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 1 | 1 | ||
| Reasoning | 3 | 2 | ||
| Math | 1 | 1 | ||
| Knowledge | 4 | 0 | ||
| General | 14 | 1 | ||
| All shared boards | 25 | 11 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index25
| Benchmark | Mistral Medium 3.5 | Hy3 preview | Leader |
|---|---|---|---|
| LMArena WebDev | 1264 | 1357 | Hy3 preview |
| AA Coding Index | 46.9 | 58.8 | Hy3 preview |
| SWE-bench Verified (Vals AI) | 66.4% | Not measured | Only Mistral Medium 3.5 measured |
| Vibe Code Bench (Vals AI) | 2.9% | Not measured | Only Mistral Medium 3.5 measured |
| Code migration (Vals AI) | 5.1% | Not measured | Only Mistral Medium 3.5 measured |
| Terminal-Bench 2.1 (Vals AI) | 39.0% | Not measured | Only Mistral Medium 3.5 measured |
| SciCode | 40.2% | Not measured | Only Mistral Medium 3.5 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 33.3% | 34.1% | Hy3 preview |
| ALE-Bench | 764.0 | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Text: coding | 1460 | 1421 | Mistral Medium 3.5 |
| LMArena Text: software and IT | 1458 | 1423 | Mistral Medium 3.5 |
| LMArena WebDev: React | 1254 | 1349 | Hy3 preview |
| LMArena WebDev: HTML | 1308 | 1356 | Hy3 preview |
| Neither model has a result on the other 26 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Mistral Medium 3.5 | Hy3 preview | Leader |
|---|---|---|---|
| AA Intelligence Index | 14.2 | 25.3 | Hy3 preview |
| Epoch Capabilities Index | 141.4 | Not measured | Only Mistral Medium 3.5 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | Mistral Medium 3.5 | Hy3 preview | Leader |
|---|---|---|---|
| AA Agentic Index | 7.0 | 24.1 | Hy3 preview |
| Finance Agent (Vals AI) | 32.1% | Not measured | Only Mistral Medium 3.5 measured |
| Tax Agent Bench (Vals AI) | 4.5% | Not measured | Only Mistral Medium 3.5 measured |
| Legal Research Bench (Vals AI) | 9.1% | Not measured | Only Mistral Medium 3.5 measured |
| Excel Modeling Benchmark (Vals AI) | 11.2% | Not measured | Only Mistral Medium 3.5 measured |
| τ²-Bench Telecom | 94.2% | 92.7% | Mistral Medium 3.5 |
| Neither model has a result on the other 35 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index32
| Benchmark | Mistral Medium 3.5 | Hy3 preview | Leader |
|---|---|---|---|
| GPQA Diamond | 34.9% | Not measured | Only Mistral Medium 3.5 measured |
| GPQA Diamond (Artificial Analysis run) | 74.8% | 86.7% | Hy3 preview |
| Humanity's Last Exam (Artificial Analysis run) | 13.8% | 27.8% | Hy3 preview |
| LMArena Text: hard prompts | 1436 | 1413 | Mistral Medium 3.5 |
| LMArena Text: expert | 1432 | 1422 | Mistral Medium 3.5 |
| LMArena Text: hard prompts in English | 1445 | 1416 | Mistral Medium 3.5 |
| LMArena Vision: diagrams | 1254 | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Vision: homework | 1230 | Not measured | Only Mistral Medium 3.5 measured |
| Kagi LLM Benchmark | 41.4% | Not measured | Only Mistral Medium 3.5 measured |
| Neither model has a result on the other 13 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | Mistral Medium 3.5 | Hy3 preview | Leader |
|---|---|---|---|
| ProofBench | 9.0% | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Text: math | 1431 | 1424 | Mistral Medium 3.5 |
| LMArena Text: mathematical work | 1440 | 1442 | Hy3 preview |
| Neither model has a result on the other 10 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | Mistral Medium 3.5 | Hy3 preview | Leader |
|---|---|---|---|
| MMLU Pro | 75.3% | Not measured | Only Mistral Medium 3.5 measured |
| CaseLaw (Vals AI) | 44.2% | Not measured | Only Mistral Medium 3.5 measured |
| CorpFin (Vals AI) | 58.8% | Not measured | Only Mistral Medium 3.5 measured |
| TaxEval (Vals AI) | 68.0% | Not measured | Only Mistral Medium 3.5 measured |
| MortgageTax (Vals AI) | 28.9% | Not measured | Only Mistral Medium 3.5 measured |
| MedCode (Vals AI) | 33.8% | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Text: science | 1428 | 1416 | Mistral Medium 3.5 |
| LMArena Text: medicine and healthcare | 1409 | 1405 | Mistral Medium 3.5 |
| LMArena Text: legal and government | 1437 | 1430 | Mistral Medium 3.5 |
| LMArena Text: business and finance | 1426 | 1397 | Mistral Medium 3.5 |
| Neither model has a result on the other 8 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | Mistral Medium 3.5 | Hy3 preview | Leader |
|---|---|---|---|
| Design Arena: websites | 1142 | Not measured | Only Mistral Medium 3.5 measured |
| Design Arena: UI components | 1079 | Not measured | Only Mistral Medium 3.5 measured |
| Design Arena: game dev | 1079 | Not measured | Only Mistral Medium 3.5 measured |
| Design Arena: data viz | 1120 | Not measured | Only Mistral Medium 3.5 measured |
| Design Arena: 3D | 1061 | Not measured | Only Mistral Medium 3.5 measured |
| Design Arena: all code categories | 1121 | Not measured | Only Mistral Medium 3.5 measured |
| Design Arena: SVG | 950 | Not measured | Only Mistral Medium 3.5 measured |
| Design Arena: ASCII art | 1063 | Not measured | 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 index141
| Benchmark | Mistral Medium 3.5 | Hy3 preview | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 37.6% | Not measured | Only Mistral Medium 3.5 measured |
| MedScribe (Vals AI) | 67.7% | Not measured | Only Mistral Medium 3.5 measured |
| Vals Multimodal Index | 34.8% | Not measured | Only Mistral Medium 3.5 measured |
| IFBench | 68.8% | 63.1% | Mistral Medium 3.5 |
| Long context reasoning (AA-LCR) | 69.3% | 64.7% | Mistral Medium 3.5 |
| DTBench | 75.5% | Not measured | Only Mistral Medium 3.5 measured |
| LMCA | 26.1% | Not measured | Only Mistral Medium 3.5 measured |
| Surface Evolver Bench | 26.9% | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Text | 1421 | 1401 | Mistral Medium 3.5 |
| LMArena Text: instruction following | 1415 | 1383 | Mistral Medium 3.5 |
| LMArena Text: creative writing | 1374 | 1339 | Mistral Medium 3.5 |
| LMArena Vision | 1222 | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Text: English | 1433 | 1409 | Mistral Medium 3.5 |
| LMArena Text: non-English | 1404 | 1390 | Mistral Medium 3.5 |
| LMArena Text: Chinese | 1444 | 1435 | Mistral Medium 3.5 |
| LMArena Text: German | 1453 | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Text: Korean | 1384 | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Text: French | 1448 | 1425 | Mistral Medium 3.5 |
| LMArena Text: Spanish | 1411 | 1399 | Mistral Medium 3.5 |
| LMArena Text: Russian | 1395 | 1399 | Hy3 preview |
| LMArena Text: Polish | 1430 | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Text: multi-turn | 1423 | 1411 | Mistral Medium 3.5 |
| LMArena Text: longer queries | 1415 | 1402 | Mistral Medium 3.5 |
| LMArena Text: writing and language | 1384 | 1370 | Mistral Medium 3.5 |
| LMArena Text: entertainment and media | 1382 | 1352 | Mistral Medium 3.5 |
| LMArena Vision: English | 1240 | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Vision: Chinese | 1244 | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Vision: reading text | 1233 | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Vision: humor | 1216 | Not measured | Only Mistral Medium 3.5 measured |
| LMArena Vision: creative writing | 1198 | Not measured | Only Mistral Medium 3.5 measured |
| Neither model has a result on the other 16 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Mistral Medium 3.5 | Hy3 preview | 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
Hy3 preview costs less
Repository review
50K input + 3K output tokens
Hy3 preview costs less
Agent loop
200K input + 10K output tokens
Hy3 preview 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 3.5 | Tencent Hy3 preview |
|---|---|---|
| Maker | Mistral AI | Tencent |
| Weights | No public weights listed | Open weights |
| Released | April 30, 2026 | April 22, 2026 |
| Context window | 262K | 262K |
| Input $/Mtok | $1.50 | $0.18 |
| Output $/Mtok | $7.50 | $0.60 |
| Coding index | 25.5 | Not benchmarked |
| Independent measurements behind it | 3 | 1 |
| Cost per solved task | $6.02 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Mistral Medium 3.5 better than Hy3 preview for coding?
Only Mistral Medium 3.5 has a coding score, 25.5 on the Vibe Coding Index; Hy3 preview has none, so the two cannot be ranked on coding. Pick Hy3 preview for output tokens 12.5x cheaper ($0.60 vs $7.50 per million) and work measured by LMArena WebDev or AA Coding Index, where it leads.
Which is cheaper, Mistral Medium 3.5 or Hy3 preview?
Mistral Medium 3.5 costs $1.50 per million input tokens and $7.50 per million output tokens; Hy3 preview costs $0.18 and $0.60. Cost per solved task is known for Mistral Medium 3.5 only.
Which has the bigger context window, Mistral Medium 3.5 or Hy3 preview?
Neither: both take 262K tokens.
How many benchmarks is the Mistral Medium 3.5 vs Hy3 preview comparison based on?
36 benchmarks have a result for both models as of October 21, 2026. Mistral Medium 3.5's coding index rests on 3 independent benchmarks and Hy3 preview's on 1. Where only one model has a result, the page says so instead of naming a winner.