Devstral 2 2512 vs GPT-6 Astra
Updated October 21, 2026. 3 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
vs boards won of 3 shared
Decision reading
Pick GPT-6 Astra for coding: it scores 88.7 on the Vibe Coding Index against 53.8 for Devstral 2 2512.
Pick Devstral 2 2512 when output tokens cost 25.0x less.
Model ADevstral 2 2512Mistral AI53.8/1001 independent benchmark$1.19 per solved taskModel BGPT-6 AstraOpenAI88.7/1002 independent benchmarksNo cost per solved task
- Shared results
- 3
- Devstral 2 2512 only
- 3
- GPT-6 Astra only
- 116
- Like-for-like categories
- 3 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
GPT-6 Astra
GPT-6 Astra scores 88.7 against 53.8.
1 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
GPT-6 Astra
GPT-6 Astra leads, 52.7 against 8.6.
Same board, both measured
Agentic work
AA Agentic Index
GPT-6 Astra
GPT-6 Astra leads, 51.0 against 2.3.
Same board, both measured
Web design
Design Arena: websites
No call
Only GPT-6 Astra has a Design Arena: websites result, 1337.
Directional only
General
Vals Index
No call
Only GPT-6 Astra has a Vals Index result, 65.0%.
Directional only
Long documents
Listed context window
GPT-6 Astra
GPT-6 Astra takes 1.05M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Devstral 2 2512
Devstral 2 2512 $0.0014, GPT-6 Astra $0.035 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, GPT-6 Astra 88.7 from 2 benchmarks.
- Widest category
- Agentic work, 48.7 points apart on average over 1 shared benchmark.
- Widest single gap
- AA Agentic Index: Devstral 2 2512 2.3 against GPT-6 Astra 51.0, 48.7 points apart on a 0 to 100 scale.
- Output price
- Devstral 2 2512 $2.00 against GPT-6 Astra $50.00 per million output tokens, 25.0x apart.
- Context window
- Devstral 2 2512 takes 262K tokens, GPT-6 Astra 1.05M.
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 | GPT-6 Astra | Ties |
|---|---|---|---|---|
| Coding | 0 | 1 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 0 | 1 | ||
| All shared boards | 0 | 3 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index01
| Benchmark | Devstral 2 2512 | GPT-6 Astra | Leader |
|---|---|---|---|
| SWE-bench Verified | 53.8% | Not measured | Only Devstral 2 2512 measured |
| LiveBench coding | Not measured | 93.7% | Only GPT-6 Astra measured |
| LMArena WebDev | Not measured | 1774 | Only GPT-6 Astra measured |
| AA Coding Index | 31.3 | 76.9 | GPT-6 Astra |
| LiveCodeBench (Vals AI) | 51.8% | Not measured | Only Devstral 2 2512 measured |
| SWE-bench Verified (Vals AI) | 62.8% | Not measured | Only Devstral 2 2512 measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 59.6% | Only GPT-6 Astra measured |
| Vibe Code Bench (Vals AI) | Not measured | 89.6% | Only GPT-6 Astra measured |
| Code migration (Vals AI) | Not measured | 67.7% | Only GPT-6 Astra measured |
| ProgramBench (Vals AI) | Not measured | 5.5% | Only GPT-6 Astra measured |
| IOI (Vals AI) | Not measured | 100.0% | Only GPT-6 Astra measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 87.3% | Only GPT-6 Astra measured |
| SWE-bench Multimodal (Vals AI) | Not measured | 43.3% | Only GPT-6 Astra measured |
| Vibe Code Bench 1-100 (Vals AI) | Not measured | 27.6% | Only GPT-6 Astra measured |
| CyberBench (Vals AI) | Not measured | 41.1% | Only GPT-6 Astra measured |
| SciCode | Not measured | 56.5% | Only GPT-6 Astra measured |
| FrontierCode | Not measured | 53.3% | Only GPT-6 Astra measured |
| FrontierSWE | Not measured | 65.5% | Only GPT-6 Astra measured |
| GSO-Bench | Not measured | 79.4% | Only GPT-6 Astra measured |
| WeirdML | Not measured | 93.3% | Only GPT-6 Astra measured |
| MirrorCode | Not measured | 46.7% | Only GPT-6 Astra measured |
| ALE-Bench | Not measured | 2951.3 | Only GPT-6 Astra measured |
| LMArena Text: coding | Not measured | 1489 | Only GPT-6 Astra measured |
| LMArena Text: software and IT | Not measured | 1474 | Only GPT-6 Astra measured |
| LMArena WebDev: React | Not measured | 1798 | Only GPT-6 Astra measured |
| LMArena WebDev: HTML | Not measured | 1775 | Only GPT-6 Astra measured |
| LMArena WebDev: from an image | Not measured | 1726 | Only GPT-6 Astra measured |
| Neither model has a result on the other 11 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Devstral 2 2512 | GPT-6 Astra | Leader |
|---|---|---|---|
| AA Intelligence Index | 8.6 | 52.7 | GPT-6 Astra |
| Epoch Capabilities Index | Not measured | 168.8 | Only GPT-6 Astra measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Devstral 2 2512 | GPT-6 Astra | Leader |
|---|---|---|---|
| AA Agentic Index | 2.3 | 51.0 | GPT-6 Astra |
| Design Arena: agents building web apps | Not measured | 1372 | Only GPT-6 Astra measured |
| Design Arena: agents building full-stack apps | Not measured | 1357 | Only GPT-6 Astra measured |
| Design Arena: agents building mobile apps | Not measured | 1312 | Only GPT-6 Astra measured |
| Design Arena: agents building Godot games | Not measured | 1348 | Only GPT-6 Astra measured |
| Design Arena: PowerPoint slides | Not measured | 1245 | Only GPT-6 Astra measured |
| Design Arena: spreadsheets | Not measured | 1380 | Only GPT-6 Astra measured |
| LiveBench research | Not measured | 85.2% | Only GPT-6 Astra measured |
| LiveBench browser | Not measured | 84.0% | Only GPT-6 Astra measured |
| LiveBench long agentic | Not measured | 82.4% | Only GPT-6 Astra measured |
| SREBench (Vals AI) | Not measured | 56.9% | Only GPT-6 Astra measured |
| Finance Agent (Vals AI) | Not measured | 53.5% | Only GPT-6 Astra measured |
| Tax Agent Bench (Vals AI) | Not measured | 20.7% | Only GPT-6 Astra measured |
| Legal Research Bench (Vals AI) | Not measured | 39.4% | Only GPT-6 Astra measured |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 7.1% | Only GPT-6 Astra measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 71.7% | Only GPT-6 Astra measured |
| BioMysteryBench (Vals AI) | Not measured | 79.3% | Only GPT-6 Astra measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 62.9% | Only GPT-6 Astra measured |
| Vals Science Index | Not measured | 65.1% | Only GPT-6 Astra measured |
| Time Horizon Index (Vals AI) | Not measured | 90.5% | Only GPT-6 Astra measured |
| CUA-bench (Vals AI) | Not measured | 19.2% | Only GPT-6 Astra measured |
| Robo-bench (Vals AI) | Not measured | 23.3% | Only GPT-6 Astra measured |
| APEX-Agents | Not measured | 64.7% | Only GPT-6 Astra measured |
| Remote Labor Index | Not measured | 20.8% | Only GPT-6 Astra measured |
| BALROG | Not measured | 68.3% | Only GPT-6 Astra measured |
| SpireClimb | Not measured | 25.7% | Only GPT-6 Astra measured |
| Vending-Bench 2 | Not measured | $15,515 | Only GPT-6 Astra measured |
| Neither model has a result on the other 12 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index15 boards, none shared
| Benchmark | Devstral 2 2512 | GPT-6 Astra | Leader |
|---|---|---|---|
| LiveBench reasoning | Not measured | 92.7% | Only GPT-6 Astra measured |
| MysteryMechanism (Vals AI) | Not measured | 53.1% | Only GPT-6 Astra measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 96.3% | Only GPT-6 Astra measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 54.7% | Only GPT-6 Astra measured |
| GPQA Diamond (Epoch AI run) | Not measured | 95.8% | Only GPT-6 Astra measured |
| ARC-AGI | Not measured | 98.5% | Only GPT-6 Astra measured |
| ARC-AGI-2 | Not measured | 95.0% | Only GPT-6 Astra measured |
| Chess puzzles | Not measured | 74.0% | Only GPT-6 Astra measured |
| Mystery game puzzles | Not measured | 84.0% | Only GPT-6 Astra measured |
| EBR-bench | Not measured | 76.2% | Only GPT-6 Astra measured |
| LMArena Text: hard prompts | Not measured | 1463 | Only GPT-6 Astra measured |
| LMArena Text: expert | Not measured | 1489 | Only GPT-6 Astra measured |
| LMArena Text: hard prompts in English | Not measured | 1464 | Only GPT-6 Astra measured |
| LMArena Vision: diagrams | Not measured | 1297 | Only GPT-6 Astra measured |
| LMArena Vision: homework | Not measured | 1337 | Only GPT-6 Astra measured |
| Neither model has a result on the other 6 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Devstral 2 2512 | GPT-6 Astra | Leader |
|---|---|---|---|
| OTIS Mock AIME | Not measured | 100.0% | Only GPT-6 Astra measured |
| FrontierMath, tiers 1 to 3 | Not measured | 93.7% | Only GPT-6 Astra measured |
| FrontierMath, tier 4 | Not measured | 97.6% | Only GPT-6 Astra measured |
| ProofBench | Not measured | 99.0% | Only GPT-6 Astra measured |
| LMArena Text: math | Not measured | 1465 | Only GPT-6 Astra measured |
| LMArena Text: mathematical work | Not measured | 1484 | Only GPT-6 Astra measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Devstral 2 2512 | GPT-6 Astra | Leader |
|---|---|---|---|
| MedCode (Vals AI) | Not measured | 48.5% | Only GPT-6 Astra measured |
| SimpleQA Verified | Not measured | 75.6% | Only GPT-6 Astra measured |
| LMArena Text: science | Not measured | 1451 | Only GPT-6 Astra measured |
| LMArena Text: medicine and healthcare | Not measured | 1425 | Only GPT-6 Astra measured |
| LMArena Text: legal and government | Not measured | 1462 | Only GPT-6 Astra measured |
| LMArena Text: business and finance | Not measured | 1439 | Only GPT-6 Astra measured |
| Neither model has a result on the other 11 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | Devstral 2 2512 | GPT-6 Astra | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1337 | Only GPT-6 Astra measured |
| Design Arena: UI components | Not measured | 1394 | Only GPT-6 Astra measured |
| Design Arena: game dev | Not measured | 1474 | Only GPT-6 Astra measured |
| Design Arena: data viz | Not measured | 1335 | Only GPT-6 Astra measured |
| Design Arena: 3D | Not measured | 1501 | Only GPT-6 Astra measured |
| Design Arena: all code categories | Not measured | 1379 | Only GPT-6 Astra measured |
| Design Arena: SVG | Not measured | 1431 | Only GPT-6 Astra measured |
| Design Arena: ASCII art | Not measured | 1359 | Only GPT-6 Astra measured |
| LiveBench design | Not measured | 82.0% | Only GPT-6 Astra measured |
GeneralDirectional onlyPublic boards, shown as published, kept out of the index30 boards, none shared
| Benchmark | Devstral 2 2512 | GPT-6 Astra | Leader |
|---|---|---|---|
| Vals Index | Not measured | 65.0% | Only GPT-6 Astra measured |
| LiveBench data analysis | Not measured | 83.0% | Only GPT-6 Astra measured |
| SAGE (Vals AI) | Not measured | 46.4% | Only GPT-6 Astra measured |
| MedScribe (Vals AI) | Not measured | 87.9% | Only GPT-6 Astra measured |
| Long context reasoning (AA-LCR) | Not measured | 80.7% | Only GPT-6 Astra measured |
| DTBench | Not measured | 97.3% | Only GPT-6 Astra measured |
| LMCA | Not measured | 64.4% | Only GPT-6 Astra measured |
| Furniture assembly | Not measured | 80.0% | Only GPT-6 Astra measured |
| LMArena Text | Not measured | 1441 | Only GPT-6 Astra measured |
| LMArena Text: instruction following | Not measured | 1455 | Only GPT-6 Astra measured |
| LMArena Text: creative writing | Not measured | 1420 | Only GPT-6 Astra measured |
| LMArena Vision | Not measured | 1280 | Only GPT-6 Astra measured |
| LMArena Text: English | Not measured | 1447 | Only GPT-6 Astra measured |
| LMArena Text: non-English | Not measured | 1432 | Only GPT-6 Astra measured |
| LMArena Text: Chinese | Not measured | 1489 | Only GPT-6 Astra measured |
| LMArena Text: German | Not measured | 1459 | Only GPT-6 Astra measured |
| LMArena Text: Japanese | Not measured | 1405 | Only GPT-6 Astra measured |
| LMArena Text: Korean | Not measured | 1428 | Only GPT-6 Astra measured |
| LMArena Text: French | Not measured | 1432 | Only GPT-6 Astra measured |
| LMArena Text: Spanish | Not measured | 1422 | Only GPT-6 Astra measured |
| LMArena Text: Russian | Not measured | 1438 | Only GPT-6 Astra measured |
| LMArena Text: Polish | Not measured | 1422 | Only GPT-6 Astra measured |
| LMArena Text: multi-turn | Not measured | 1447 | Only GPT-6 Astra measured |
| LMArena Text: longer queries | Not measured | 1460 | Only GPT-6 Astra measured |
| LMArena Text: writing and language | Not measured | 1437 | Only GPT-6 Astra measured |
| LMArena Text: entertainment and media | Not measured | 1427 | Only GPT-6 Astra measured |
| LMArena Vision: English | Not measured | 1280 | Only GPT-6 Astra measured |
| LMArena Vision: Chinese | Not measured | 1361 | Only GPT-6 Astra measured |
| LMArena Vision: reading text | Not measured | 1292 | Only GPT-6 Astra measured |
| LMArena Vision: creative writing | Not measured | 1286 | Only GPT-6 Astra 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 | Devstral 2 2512 | GPT-6 Astra | 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$0.0014Fits in one request
GPT-6 Astra$0.035Fits in one request
Devstral 2 2512 costs less
Repository review
50K input + 3K output tokens
Devstral 2 2512$0.026Fits in one request
GPT-6 Astra$0.650Fits in one request
Devstral 2 2512 costs less
Agent loop
200K input + 10K output tokens
Devstral 2 2512$0.100Fits in one request
GPT-6 Astra$2.50Fits in one request
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 | OpenAI GPT-6 Astra |
|---|---|---|
| Maker | Mistral AI | OpenAI |
| Weights | Open weights | No public weights listed |
| Released | December 9, 2025 | September 4, 2026 |
| Context window | 262K | 1.05M |
| Input $/Mtok | $0.40 | $10.00 |
| Output $/Mtok | $2.00 | $50.00 |
| Coding index | 53.8 | 88.7 |
| 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 GPT-6 Astra for coding?
Pick GPT-6 Astra for coding: it scores 88.7 on the Vibe Coding Index against 53.8 for Devstral 2 2512. Pick Devstral 2 2512 when output tokens cost 25.0x less.
Which is cheaper, Devstral 2 2512 or GPT-6 Astra?
Devstral 2 2512 costs $0.40 per million input tokens and $2.00 per million output tokens; GPT-6 Astra costs $10.00 and $50.00. Cost per solved task is known for Devstral 2 2512 only.
Which has the bigger context window, Devstral 2 2512 or GPT-6 Astra?
GPT-6 Astra, with 1.05M tokens against 262K.
How many benchmarks is the Devstral 2 2512 vs GPT-6 Astra comparison based on?
3 benchmarks have a result for both models as of October 21, 2026. Devstral 2 2512's coding index rests on 1 independent benchmark and GPT-6 Astra's on 2. Where only one model has a result, the page says so instead of naming a winner.