Devstral 2 2512 vs GLM 4.6
Updated October 21, 2026. 4 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Devstral 2 2512 and GLM 4.6 are level for coding, 53.8 against 52.2 on the Vibe Coding Index, so price and context decide.
Pick GLM 4.6 when what matters is SWE-bench Verified or AA Coding Index, where it leads. Pick Devstral 2 2512 when output tokens cost 1.1x less, or when the context window is larger, 262K against 205K tokens.
- Shared results
- 4
- Devstral 2 2512 only
- 2
- GLM 4.6 only
- 62
- Like-for-like categories
- 2 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, four public coding benchmarks
No call
Level: 53.8 against 52.2, inside 2 points.
1 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
GLM 4.6
GLM 4.6 leads, 18.5 against 8.6.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Devstral 2 2512 has a AA Agentic Index result, 2.3.
Directional only
Reasoning
GPQA Diamond
No call
Only GLM 4.6 has a GPQA Diamond result, 74.5%.
Directional only
Math
AIME
No call
Only GLM 4.6 has a AIME result, 92.7%.
Directional only
Knowledge
MMLU Pro
No call
Only GLM 4.6 has a MMLU Pro result, 82.2%.
Directional only
Web design
Design Arena: websites
No call
Only GLM 4.6 has a Design Arena: websites result, 1174.
Directional only
Long documents
Listed context window
Devstral 2 2512
Devstral 2 2512 takes 262K tokens against 205K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Devstral 2 2512
Devstral 2 2512 $0.0014, GLM 4.6 $0.0016 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Devstral 2 2512 has the number, $1.19.
Directional only
At a glance
- Coding index
- Devstral 2 2512 53.8 from 1 benchmark, GLM 4.6 52.2 from 2 benchmarks.
- Widest category
- Coding, 19.4 points apart on average over 3 shared benchmarks.
- Widest single gap
- LiveCodeBench (Vals AI): Devstral 2 2512 51.8% against GLM 4.6 81.0%, 29.2 points apart on a 0 to 100 scale.
- Output price
- Devstral 2 2512 $2.00 against GLM 4.6 $2.20 per million output tokens, 1.1x apart.
- Context window
- Devstral 2 2512 takes 262K tokens, GLM 4.6 205K.
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 | Devstral 2 2512 | Bars | GLM 4.6 | Ties |
|---|---|---|---|---|
| Coding | 0 | 3 | ||
| Intelligence | 0 | 1 | ||
| All shared boards | 0 | 4 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index03
| Benchmark | Devstral 2 2512 | GLM 4.6 | Leader |
|---|---|---|---|
| SWE-bench Verified | 53.8% | 68.2% | GLM 4.6 |
| LMArena WebDev | Not measured | 1341 | Only GLM 4.6 measured |
| AA Coding Index | 31.3 | 45.8 | GLM 4.6 |
| LiveCodeBench (Vals AI) | 51.8% | 81.0% | GLM 4.6 |
| SWE-bench Verified (Vals AI) | 62.8% | Not measured | Only Devstral 2 2512 measured |
| Vibe Code Bench (Vals AI) | Not measured | 3.1% | Only GLM 4.6 measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 69.5% | Only GLM 4.6 measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 28.8% | Only GLM 4.6 measured |
| Terminal-Bench | Not measured | 24.5% | Only GLM 4.6 measured |
| ALE-Bench | Not measured | 340.8 | Only GLM 4.6 measured |
| LMArena Text: coding | Not measured | 1449 | Only GLM 4.6 measured |
| LMArena Text: software and IT | Not measured | 1455 | Only GLM 4.6 measured |
| LMArena WebDev: HTML | Not measured | 1350 | Only GLM 4.6 measured |
| Neither model has a result on the other 25 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Devstral 2 2512 | GLM 4.6 | Leader |
|---|---|---|---|
| AA Intelligence Index | 8.6 | 18.5 | GLM 4.6 |
| Neither model has a result on the other 1 board here | |||
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Devstral 2 2512 | GLM 4.6 | Leader |
|---|---|---|---|
| AA Agentic Index | 2.3 | Not measured | Only Devstral 2 2512 measured |
| Design Arena: agents building full-stack apps | Not measured | 1029 | Only GLM 4.6 measured |
| Design Arena: agents building mobile apps | Not measured | 1096 | Only GLM 4.6 measured |
| Design Arena: agents building Android apps | Not measured | 1090 | Only GLM 4.6 measured |
| Design Arena: agents building Godot games | Not measured | 1180 | Only GLM 4.6 measured |
| τ²-Bench Telecom | Not measured | 76.9% | Only GLM 4.6 measured |
| Neither model has a result on the other 33 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Devstral 2 2512 | GLM 4.6 | Leader |
|---|---|---|---|
| GPQA Diamond | Not measured | 74.5% | Only GLM 4.6 measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 78.0% | Only GLM 4.6 measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 14.5% | Only GLM 4.6 measured |
| LMArena Text: hard prompts | Not measured | 1440 | Only GLM 4.6 measured |
| LMArena Text: expert | Not measured | 1432 | Only GLM 4.6 measured |
| LMArena Text: hard prompts in English | Not measured | 1445 | Only GLM 4.6 measured |
| Neither model has a result on the other 15 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Devstral 2 2512 | GLM 4.6 | Leader |
|---|---|---|---|
| AIME | Not measured | 92.7% | Only GLM 4.6 measured |
| MGSM (Vals AI) | Not measured | 89.8% | Only GLM 4.6 measured |
| AA Math Index | Not measured | 86.0 | Only GLM 4.6 measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 86.0% | Only GLM 4.6 measured |
| LMArena Text: math | Not measured | 1432 | Only GLM 4.6 measured |
| LMArena Text: mathematical work | Not measured | 1447 | Only GLM 4.6 measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index10 boards, none shared
| Benchmark | Devstral 2 2512 | GLM 4.6 | Leader |
|---|---|---|---|
| MMLU Pro | Not measured | 82.2% | Only GLM 4.6 measured |
| LegalBench (Vals AI) | Not measured | 79.6% | Only GLM 4.6 measured |
| CorpFin (Vals AI) | Not measured | 56.8% | Only GLM 4.6 measured |
| TaxEval (Vals AI) | Not measured | 66.2% | Only GLM 4.6 measured |
| MedQA (Vals AI) | Not measured | 92.2% | Only GLM 4.6 measured |
| MMLU Pro (Artificial Analysis run) | Not measured | 82.9% | Only GLM 4.6 measured |
| LMArena Text: science | Not measured | 1456 | Only GLM 4.6 measured |
| LMArena Text: medicine and healthcare | Not measured | 1455 | Only GLM 4.6 measured |
| LMArena Text: legal and government | Not measured | 1448 | Only GLM 4.6 measured |
| LMArena Text: business and finance | Not measured | 1416 | Only GLM 4.6 measured |
| Neither model has a result on the other 7 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
| Benchmark | Devstral 2 2512 | GLM 4.6 | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1174 | Only GLM 4.6 measured |
| Design Arena: UI components | Not measured | 1147 | Only GLM 4.6 measured |
| Design Arena: game dev | Not measured | 1151 | Only GLM 4.6 measured |
| Design Arena: data viz | Not measured | 1161 | Only GLM 4.6 measured |
| Design Arena: 3D | Not measured | 1129 | Only GLM 4.6 measured |
| Design Arena: all code categories | Not measured | 1166 | Only GLM 4.6 measured |
| Design Arena: SVG | Not measured | 1108 | Only GLM 4.6 measured |
| Neither model has a result on the other 2 boards here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index19 boards, none shared
| Benchmark | Devstral 2 2512 | GLM 4.6 | Leader |
|---|---|---|---|
| IFBench | Not measured | 43.4% | Only GLM 4.6 measured |
| Long context reasoning (AA-LCR) | Not measured | 54.0% | Only GLM 4.6 measured |
| LMArena Text | Not measured | 1440 | Only GLM 4.6 measured |
| LMArena Text: instruction following | Not measured | 1410 | Only GLM 4.6 measured |
| LMArena Text: creative writing | Not measured | 1412 | Only GLM 4.6 measured |
| LMArena Text: English | Not measured | 1449 | Only GLM 4.6 measured |
| LMArena Text: non-English | Not measured | 1426 | Only GLM 4.6 measured |
| LMArena Text: Chinese | Not measured | 1499 | Only GLM 4.6 measured |
| LMArena Text: German | Not measured | 1447 | Only GLM 4.6 measured |
| LMArena Text: Japanese | Not measured | 1393 | Only GLM 4.6 measured |
| LMArena Text: Korean | Not measured | 1400 | Only GLM 4.6 measured |
| LMArena Text: French | Not measured | 1459 | Only GLM 4.6 measured |
| LMArena Text: Spanish | Not measured | 1436 | Only GLM 4.6 measured |
| LMArena Text: Russian | Not measured | 1418 | Only GLM 4.6 measured |
| LMArena Text: Polish | Not measured | 1447 | Only GLM 4.6 measured |
| LMArena Text: multi-turn | Not measured | 1427 | Only GLM 4.6 measured |
| LMArena Text: longer queries | Not measured | 1422 | Only GLM 4.6 measured |
| LMArena Text: writing and language | Not measured | 1415 | Only GLM 4.6 measured |
| LMArena Text: entertainment and media | Not measured | 1409 | Only GLM 4.6 measured |
| Neither model has a result on the other 27 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Devstral 2 2512 | GLM 4.6 | 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
Devstral 2 2512 costs less
Repository review
50K input + 3K output tokens
Devstral 2 2512 costs less
Agent loop
200K input + 10K output tokens
Devstral 2 2512 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 Devstral 2 2512 | Z.ai GLM 4.6 |
|---|---|---|
| Maker | Mistral AI | Z.ai |
| Weights | Open weights | Open weights |
| Released | December 9, 2025 | September 30, 2025 |
| Context window | 262K | 205K |
| Input $/Mtok | $0.40 | $0.55 |
| Output $/Mtok | $2.00 | $2.20 |
| Coding index | 53.8 | 52.2 |
| Independent measurements behind it | 1 | 2 |
| Cost per solved task | $1.19 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Devstral 2 2512 better than GLM 4.6 for coding?
Devstral 2 2512 and GLM 4.6 are level for coding, 53.8 against 52.2 on the Vibe Coding Index, so price and context decide. Pick GLM 4.6 when what matters is SWE-bench Verified or AA Coding Index, where it leads. Pick Devstral 2 2512 when output tokens cost 1.1x less, or when the context window is larger, 262K against 205K tokens.
Which is cheaper, Devstral 2 2512 or GLM 4.6?
Devstral 2 2512 costs $0.40 per million input tokens and $2.00 per million output tokens; GLM 4.6 costs $0.55 and $2.20. Cost per solved task is known for Devstral 2 2512 only.
Which has the bigger context window, Devstral 2 2512 or GLM 4.6?
Devstral 2 2512, with 262K tokens against 205K.
How many benchmarks is the Devstral 2 2512 vs GLM 4.6 comparison based on?
4 benchmarks have a result for both models as of October 21, 2026. Devstral 2 2512's coding index rests on 1 independent benchmark and GLM 4.6's on 2. Where only one model has a result, the page says so instead of naming a winner.