Codestral 2508 vs Qwen3.5 397B A17B
Codestral 2508 costs 2.6x less per output token than Qwen3.5 397B A17B ($0.90 vs $2.34 per million). Updated October 21, 2026; 7 benchmarks have a result for both.
Codestral 2508 or Qwen3.5 397B A17B: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Codestral 2508 for output tokens 2.6x cheaper ($0.90 vs $2.34 per million) and work measured by ALE-Bench or Design Arena: websites, where it leads.
- Shared results
- 7
- Codestral 2508 only
- 1
- Qwen3.5 397B A17B only
- 58
- Like-for-like categories
- 2 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
No call
Only Qwen3.5 397B A17B has a AA Intelligence Index result, 21.4.
Directional only
Web design
Design Arena: websites
Qwen3.5 397B A17B
Qwen3.5 397B A17B leads, 1191 against 1013.
Same board, both measured
Long documents
Listed context window
Qwen3.5 397B A17B
Qwen3.5 397B A17B takes 262K tokens against 256K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Codestral 2508
Codestral 2508 $0.0008, Qwen3.5 397B A17B $0.0016 per turn.
List prices
At a glance
- Coding index
- Codestral 2508 not benchmarked, Qwen3.5 397B A17B not benchmarked.
- Widest category
- Web design, 20.2 points apart on average over 6 shared benchmarks.
- Widest single gap
- Design Arena: websites: Codestral 2508 1013 against Qwen3.5 397B A17B 1191, 22.3 points apart on a 0 to 100 scale.
- Output price
- Codestral 2508 $0.90 against Qwen3.5 397B A17B $2.34 per million output tokens, 2.6x apart.
- Context window
- Codestral 2508 takes 256K tokens, Qwen3.5 397B A17B 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 | Codestral 2508 | Bars | Qwen3.5 397B A17B | Ties |
|---|---|---|---|---|
| Coding | 0 | 1 | ||
| Web design | 0 | 6 | ||
| All shared boards | 0 | 7 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index01
| Benchmark | Codestral 2508 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| LMArena WebDev | Not measured | 1400 | Only Qwen3.5 397B A17B measured |
| SciCode | Not measured | 44.8% | Only Qwen3.5 397B A17B measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 40.9% | Only Qwen3.5 397B A17B measured |
| ALE-Bench | 137.8 | 621.9 | Qwen3.5 397B A17B |
| LMArena Text: coding | Not measured | 1464 | Only Qwen3.5 397B A17B measured |
| LMArena Text: software and IT | Not measured | 1463 | Only Qwen3.5 397B A17B measured |
| LMArena WebDev: React | Not measured | 1388 | Only Qwen3.5 397B A17B measured |
| LMArena WebDev: HTML | Not measured | 1410 | Only Qwen3.5 397B A17B measured |
| Neither model has a result on the other 31 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Codestral 2508 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| AA Intelligence Index | Not measured | 21.4 | Only Qwen3.5 397B A17B measured |
| Epoch Capabilities Index | Not measured | 146.8 | Only Qwen3.5 397B A17B measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Codestral 2508 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| τ²-Bench Telecom | Not measured | 95.6% | Only Qwen3.5 397B A17B measured |
| APEX-Agents | Not measured | 24.9% | Only Qwen3.5 397B A17B measured |
| Neither model has a result on the other 39 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index10 boards, none shared
| Benchmark | Codestral 2508 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | Not measured | 89.3% | Only Qwen3.5 397B A17B measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 29.0% | Only Qwen3.5 397B A17B measured |
| GPQA Diamond (Epoch AI run) | Not measured | 86.4% | Only Qwen3.5 397B A17B measured |
| Mystery game puzzles | Not measured | 18.0% | Only Qwen3.5 397B A17B measured |
| LMArena Text: hard prompts | Not measured | 1448 | Only Qwen3.5 397B A17B measured |
| LMArena Text: expert | Not measured | 1462 | Only Qwen3.5 397B A17B measured |
| LMArena Text: hard prompts in English | Not measured | 1450 | Only Qwen3.5 397B A17B measured |
| LMArena Vision: diagrams | Not measured | 1277 | Only Qwen3.5 397B A17B measured |
| LMArena Vision: homework | Not measured | 1292 | Only Qwen3.5 397B A17B measured |
| Kagi LLM Benchmark | 32.5% | Not measured | Only Codestral 2508 measured |
| Neither model has a result on the other 12 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
| Benchmark | Codestral 2508 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| OTIS Mock AIME | Not measured | 88.9% | Only Qwen3.5 397B A17B measured |
| FrontierMath, tiers 1 to 3 | Not measured | 31.2% | Only Qwen3.5 397B A17B measured |
| LMArena Text: math | Not measured | 1453 | Only Qwen3.5 397B A17B measured |
| LMArena Text: mathematical work | Not measured | 1454 | Only Qwen3.5 397B A17B measured |
| Neither model has a result on the other 9 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
| Benchmark | Codestral 2508 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| LMArena Text: science | Not measured | 1457 | Only Qwen3.5 397B A17B measured |
| LMArena Text: medicine and healthcare | Not measured | 1455 | Only Qwen3.5 397B A17B measured |
| LMArena Text: legal and government | Not measured | 1441 | Only Qwen3.5 397B A17B measured |
| LMArena Text: business and finance | Not measured | 1435 | Only Qwen3.5 397B A17B measured |
| Neither model has a result on the other 14 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | Codestral 2508 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| Design Arena: websites | 1013 | 1191 | Qwen3.5 397B A17B |
| Design Arena: UI components | 1017 | 1163 | Qwen3.5 397B A17B |
| Design Arena: game dev | 968 | 1140 | Qwen3.5 397B A17B |
| Design Arena: data viz | 1017 | 1175 | Qwen3.5 397B A17B |
| Design Arena: 3D | 1020 | 1161 | Qwen3.5 397B A17B |
| Design Arena: all code categories | 1007 | 1180 | Qwen3.5 397B A17B |
| Design Arena: SVG | Not measured | 1140 | Only Qwen3.5 397B A17B measured |
| Neither model has a result on the other 2 boards here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index29 boards, none shared
| Benchmark | Codestral 2508 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| IFBench | Not measured | 78.8% | Only Qwen3.5 397B A17B measured |
| Long context reasoning (AA-LCR) | Not measured | 77.3% | Only Qwen3.5 397B A17B measured |
| DTBench | Not measured | 87.5% | Only Qwen3.5 397B A17B measured |
| LMCA | Not measured | 37.9% | Only Qwen3.5 397B A17B measured |
| LMArena Text | Not measured | 1437 | Only Qwen3.5 397B A17B measured |
| LMArena Text: instruction following | Not measured | 1424 | Only Qwen3.5 397B A17B measured |
| LMArena Text: creative writing | Not measured | 1402 | Only Qwen3.5 397B A17B measured |
| LMArena Vision | Not measured | 1262 | Only Qwen3.5 397B A17B measured |
| LMArena Text: English | Not measured | 1443 | Only Qwen3.5 397B A17B measured |
| LMArena Text: non-English | Not measured | 1430 | Only Qwen3.5 397B A17B measured |
| LMArena Text: Chinese | Not measured | 1498 | Only Qwen3.5 397B A17B measured |
| LMArena Text: German | Not measured | 1445 | Only Qwen3.5 397B A17B measured |
| LMArena Text: Japanese | Not measured | 1429 | Only Qwen3.5 397B A17B measured |
| LMArena Text: Korean | Not measured | 1384 | Only Qwen3.5 397B A17B measured |
| LMArena Text: French | Not measured | 1462 | Only Qwen3.5 397B A17B measured |
| LMArena Text: Spanish | Not measured | 1441 | Only Qwen3.5 397B A17B measured |
| LMArena Text: Russian | Not measured | 1430 | Only Qwen3.5 397B A17B measured |
| LMArena Text: Polish | Not measured | 1447 | Only Qwen3.5 397B A17B measured |
| LMArena Text: multi-turn | Not measured | 1446 | Only Qwen3.5 397B A17B measured |
| LMArena Text: longer queries | Not measured | 1441 | Only Qwen3.5 397B A17B measured |
| LMArena Text: writing and language | Not measured | 1415 | Only Qwen3.5 397B A17B measured |
| LMArena Text: entertainment and media | Not measured | 1399 | Only Qwen3.5 397B A17B measured |
| LMArena Vision: English | Not measured | 1268 | Only Qwen3.5 397B A17B measured |
| LMArena Vision: Chinese | Not measured | 1314 | Only Qwen3.5 397B A17B measured |
| LMArena Vision: reading text | Not measured | 1271 | Only Qwen3.5 397B A17B measured |
| LMArena Vision: humor | Not measured | 1262 | Only Qwen3.5 397B A17B measured |
| LMArena Vision: captioning | Not measured | 1230 | Only Qwen3.5 397B A17B measured |
| LMArena Vision: recognizing things | Not measured | 1265 | Only Qwen3.5 397B A17B measured |
| LMArena Vision: creative writing | Not measured | 1260 | Only Qwen3.5 397B A17B 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 | Codestral 2508 | Qwen3.5 397B A17B | 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
Codestral 2508 costs less
Repository review
50K input + 3K output tokens
Codestral 2508 costs less
Agent loop
200K input + 10K output tokens
Codestral 2508 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 Codestral 2508 | Qwen3.5 397B A17B |
|---|---|---|
| Maker | Mistral AI | Qwen |
| Weights | No public weights listed | Open weights |
| Released | August 1, 2025 | February 16, 2026 |
| Context window | 256K | 262K |
| Input $/Mtok | $0.30 | $0.39 |
| Output $/Mtok | $0.90 | $2.34 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 0 | 1 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Codestral 2508 better than Qwen3.5 397B A17B for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Codestral 2508 for output tokens 2.6x cheaper ($0.90 vs $2.34 per million) and work measured by ALE-Bench or Design Arena: websites, where it leads.
Which is cheaper, Codestral 2508 or Qwen3.5 397B A17B?
Codestral 2508 costs $0.30 per million input tokens and $0.90 per million output tokens; Qwen3.5 397B A17B costs $0.39 and $2.34. Cost per solved task is known for neither of them.
Which has the bigger context window, Codestral 2508 or Qwen3.5 397B A17B?
Qwen3.5 397B A17B, with 262K tokens against 256K.
How many benchmarks is the Codestral 2508 vs Qwen3.5 397B A17B comparison based on?
7 benchmarks have a result for both models as of October 21, 2026. Codestral 2508's coding index rests on 0 independent benchmarks and Qwen3.5 397B A17B's on 1. Where only one model has a result, the page says so instead of naming a winner.