Devstral 2 2512 vs GPT-5.4 Mini
GPT-5.4 Mini leads Devstral 2 2512 on the Artificial Analysis Intelligence Index, 24.1 to 8.6, and Devstral 2 2512 costs 2.3x less per output token ($2.00 vs $4.50 per million). Updated October 22, 2026; 5 benchmarks have a result for both.
Devstral 2 2512 or GPT-5.4 Mini: which to pick
Only GPT-5.4 Mini has a coding score, 35.8 on the Vibe Coding Index; Devstral 2 2512 has none, so the two cannot be ranked on coding.
Pick Devstral 2 2512 for output tokens 2.3x cheaper ($2.00 vs $4.50 per million).
- Shared results
- 5
- Devstral 2 2512 only
- 1
- GPT-5.4 Mini only
- 84
- Like-for-like categories
- 3 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 GPT-5.4 Mini has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
GPT-5.4 Mini
GPT-5.4 Mini leads, 24.1 against 8.6.
Same board, both measured
Agentic work
AA Agentic Index
GPT-5.4 Mini
GPT-5.4 Mini leads, 17.9 against 2.3.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only GPT-5.4 Mini has a GPQA Diamond result, 83.1%.
Directional only
Math
AIME
No call
Only GPT-5.4 Mini has a AIME result, 95.6%.
Directional only
Knowledge
MMLU Pro
No call
Only GPT-5.4 Mini has a MMLU Pro result, 84.5%.
Directional only
Long documents
Listed context window
GPT-5.4 Mini
GPT-5.4 Mini takes 400K tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Devstral 2 2512
Devstral 2 2512 $0.0014, GPT-5.4 Mini $0.0030 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
GPT-5.4 Mini
Devstral 2 2512 $1.19, GPT-5.4 Mini $0.70.
Published attempt cost and pass rate
At a glance
- Coding index
- Devstral 2 2512 not benchmarked, GPT-5.4 Mini 35.8 from 5 benchmarks.
- Widest category
- Coding, 21.5 points apart on average over 3 shared benchmarks.
- Widest single gap
- LiveCodeBench (Vals AI): Devstral 2 2512 51.8% against GPT-5.4 Mini 81.5%, 29.6 points apart on a 0 to 100 scale.
- Output price
- Devstral 2 2512 $2.00 against GPT-5.4 Mini $4.50 per million output tokens, 2.3x apart.
- Context window
- Devstral 2 2512 takes 262K tokens, GPT-5.4 Mini 400K.
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-5.4 Mini | Ties |
|---|---|---|---|---|
| Coding | 0 | 3 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 0 | 1 | ||
| All shared boards | 0 | 5 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index03
| Benchmark | Devstral 2 2512 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| SWE-bench Verified | 53.8% | Not measured | Only Devstral 2 2512 measured |
| LMArena WebDev | Not measured | 1397 | Only GPT-5.4 Mini measured |
| AA Coding Index | 31.3 | 56.1 | GPT-5.4 Mini |
| LiveCodeBench (Vals AI) | 51.8% | 81.5% | GPT-5.4 Mini |
| SWE-bench Verified (Vals AI) | 62.8% | 73.0% | GPT-5.4 Mini |
| Terminal-Bench 4 (Vals AI) | Not measured | 2.5% | Only GPT-5.4 Mini measured |
| Vibe Code Bench (Vals AI) | Not measured | 48.0% | Only GPT-5.4 Mini measured |
| Code migration (Vals AI) | Not measured | 12.9% | Only GPT-5.4 Mini measured |
| ProgramBench (Vals AI) | Not measured | 0.0% | Only GPT-5.4 Mini measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 54.7% | Only GPT-5.4 Mini measured |
| SciCode | Not measured | 52.1% | Only GPT-5.4 Mini measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 52.3% | Only GPT-5.4 Mini measured |
| FrontierCode | Not measured | 27.0% | Only GPT-5.4 Mini measured |
| WeirdML | Not measured | 60.3% | Only GPT-5.4 Mini measured |
| ALE-Bench | Not measured | 1188.6 | Only GPT-5.4 Mini measured |
| LMArena Text: coding | Not measured | 1438 | Only GPT-5.4 Mini measured |
| LMArena Text: software and IT | Not measured | 1436 | Only GPT-5.4 Mini measured |
| LMArena WebDev: React | Not measured | 1386 | Only GPT-5.4 Mini measured |
| LMArena WebDev: HTML | Not measured | 1419 | Only GPT-5.4 Mini measured |
| Neither model has a result on the other 20 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Devstral 2 2512 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| AA Intelligence Index | 8.6 | 24.1 | GPT-5.4 Mini |
| Epoch Capabilities Index | Not measured | 148.7 | Only GPT-5.4 Mini measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Devstral 2 2512 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| AA Agentic Index | 2.3 | 17.9 | GPT-5.4 Mini |
| Finance Agent (Vals AI) | Not measured | 45.4% | Only GPT-5.4 Mini measured |
| Tax Agent Bench (Vals AI) | Not measured | 8.8% | Only GPT-5.4 Mini measured |
| Legal Research Bench (Vals AI) | Not measured | 12.5% | Only GPT-5.4 Mini measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 45.4% | Only GPT-5.4 Mini measured |
| τ²-Bench Telecom | Not measured | 83.3% | Only GPT-5.4 Mini measured |
| DeepResearch Bench | Not measured | 36.3% | Only GPT-5.4 Mini measured |
| Neither model has a result on the other 34 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index15 boards, none shared
| Benchmark | Devstral 2 2512 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| GPQA Diamond | Not measured | 83.1% | Only GPT-5.4 Mini measured |
| MMMU Pro (Vals AI) | Not measured | 79.3% | Only GPT-5.4 Mini measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 87.5% | Only GPT-5.4 Mini measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 28.1% | Only GPT-5.4 Mini measured |
| GPQA Diamond (Epoch AI run) | Not measured | 86.9% | Only GPT-5.4 Mini measured |
| ARC-AGI | Not measured | 63.7% | Only GPT-5.4 Mini measured |
| ARC-AGI-2 | Not measured | 18.9% | Only GPT-5.4 Mini measured |
| Chess puzzles | Not measured | 23.0% | Only GPT-5.4 Mini measured |
| Mystery game puzzles | Not measured | 11.0% | Only GPT-5.4 Mini measured |
| LMArena Text: hard prompts | Not measured | 1424 | Only GPT-5.4 Mini measured |
| LMArena Text: expert | Not measured | 1435 | Only GPT-5.4 Mini measured |
| LMArena Text: hard prompts in English | Not measured | 1420 | Only GPT-5.4 Mini measured |
| LMArena Vision: diagrams | Not measured | 1266 | Only GPT-5.4 Mini measured |
| LMArena Vision: homework | Not measured | 1285 | Only GPT-5.4 Mini measured |
| Kagi LLM Benchmark | Not measured | 37.9% | Only GPT-5.4 Mini measured |
| Neither model has a result on the other 7 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
| Benchmark | Devstral 2 2512 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| AIME | Not measured | 95.6% | Only GPT-5.4 Mini measured |
| OTIS Mock AIME | Not measured | 88.9% | Only GPT-5.4 Mini measured |
| FrontierMath, tiers 1 to 3 | Not measured | 51.2% | Only GPT-5.4 Mini measured |
| FrontierMath, tier 4 | Not measured | 9.8% | Only GPT-5.4 Mini measured |
| ProofBench | Not measured | 21.0% | Only GPT-5.4 Mini measured |
| LMArena Text: math | Not measured | 1418 | Only GPT-5.4 Mini measured |
| LMArena Text: mathematical work | Not measured | 1429 | Only GPT-5.4 Mini measured |
| Neither model has a result on the other 6 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index10 boards, none shared
| Benchmark | Devstral 2 2512 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| MMLU Pro | Not measured | 84.5% | Only GPT-5.4 Mini measured |
| CaseLaw (Vals AI) | Not measured | 51.7% | Only GPT-5.4 Mini measured |
| CorpFin (Vals AI) | Not measured | 60.9% | Only GPT-5.4 Mini measured |
| TaxEval (Vals AI) | Not measured | 71.2% | Only GPT-5.4 Mini measured |
| MortgageTax (Vals AI) | Not measured | 63.5% | Only GPT-5.4 Mini measured |
| SimpleQA Verified | Not measured | 29.4% | Only GPT-5.4 Mini measured |
| LMArena Text: science | Not measured | 1417 | Only GPT-5.4 Mini measured |
| LMArena Text: medicine and healthcare | Not measured | 1395 | Only GPT-5.4 Mini measured |
| LMArena Text: legal and government | Not measured | 1414 | Only GPT-5.4 Mini measured |
| LMArena Text: business and finance | Not measured | 1418 | Only GPT-5.4 Mini measured |
| Neither model has a result on the other 8 boards here | |||
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Devstral 2 2512 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| Neither model has a result on 9 boards here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index30 boards, none shared
| Benchmark | Devstral 2 2512 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| SAGE (Vals AI) | Not measured | 50.8% | Only GPT-5.4 Mini measured |
| Vals Multimodal Index | Not measured | 54.2% | Only GPT-5.4 Mini measured |
| IFBench | Not measured | 73.3% | Only GPT-5.4 Mini measured |
| Long context reasoning (AA-LCR) | Not measured | 77.0% | Only GPT-5.4 Mini measured |
| DTBench | Not measured | 80.0% | Only GPT-5.4 Mini measured |
| LMCA | Not measured | 40.8% | Only GPT-5.4 Mini measured |
| LMArena Text | Not measured | 1412 | Only GPT-5.4 Mini measured |
| LMArena Text: instruction following | Not measured | 1405 | Only GPT-5.4 Mini measured |
| LMArena Text: creative writing | Not measured | 1370 | Only GPT-5.4 Mini measured |
| LMArena Vision | Not measured | 1245 | Only GPT-5.4 Mini measured |
| LMArena Text: English | Not measured | 1412 | Only GPT-5.4 Mini measured |
| LMArena Text: non-English | Not measured | 1405 | Only GPT-5.4 Mini measured |
| LMArena Text: Chinese | Not measured | 1446 | Only GPT-5.4 Mini measured |
| LMArena Text: German | Not measured | 1408 | Only GPT-5.4 Mini measured |
| LMArena Text: Japanese | Not measured | 1375 | Only GPT-5.4 Mini measured |
| LMArena Text: Korean | Not measured | 1370 | Only GPT-5.4 Mini measured |
| LMArena Text: French | Not measured | 1441 | Only GPT-5.4 Mini measured |
| LMArena Text: Spanish | Not measured | 1405 | Only GPT-5.4 Mini measured |
| LMArena Text: Russian | Not measured | 1417 | Only GPT-5.4 Mini measured |
| LMArena Text: Polish | Not measured | 1415 | Only GPT-5.4 Mini measured |
| LMArena Text: multi-turn | Not measured | 1429 | Only GPT-5.4 Mini measured |
| LMArena Text: longer queries | Not measured | 1408 | Only GPT-5.4 Mini measured |
| LMArena Text: writing and language | Not measured | 1393 | Only GPT-5.4 Mini measured |
| LMArena Text: entertainment and media | Not measured | 1373 | Only GPT-5.4 Mini measured |
| LMArena Vision: English | Not measured | 1244 | Only GPT-5.4 Mini measured |
| LMArena Vision: Chinese | Not measured | 1286 | Only GPT-5.4 Mini measured |
| LMArena Vision: reading text | Not measured | 1253 | Only GPT-5.4 Mini measured |
| LMArena Vision: humor | Not measured | 1235 | Only GPT-5.4 Mini measured |
| LMArena Vision: recognizing things | Not measured | 1218 | Only GPT-5.4 Mini measured |
| LMArena Vision: creative writing | Not measured | 1222 | Only GPT-5.4 Mini 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-5.4 Mini | 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 | OpenAI GPT-5.4 Mini |
|---|---|---|
| Maker | Mistral AI | OpenAI |
| Weights | Open weights | No public weights listed |
| Released | December 9, 2025 | March 17, 2026 |
| Context window | 262K | 400K |
| Input $/Mtok | $0.40 | $0.75 |
| Output $/Mtok | $2.00 | $4.50 |
| Coding index | Not benchmarked | 35.8 |
| Independent measurements behind it | 1 | 5 |
| Cost per solved task | $1.19 | $0.70 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Devstral 2 2512 better than GPT-5.4 Mini for coding?
Only GPT-5.4 Mini has a coding score, 35.8 on the Vibe Coding Index; Devstral 2 2512 has none, so the two cannot be ranked on coding. Pick Devstral 2 2512 for output tokens 2.3x cheaper ($2.00 vs $4.50 per million).
Which is cheaper, Devstral 2 2512 or GPT-5.4 Mini?
Devstral 2 2512 costs $0.40 per million input tokens and $2.00 per million output tokens; GPT-5.4 Mini costs $0.75 and $4.50. Per solved SWE-bench Verified task, Devstral 2 2512 costs $1.19 and GPT-5.4 Mini $0.70.
Which has the bigger context window, Devstral 2 2512 or GPT-5.4 Mini?
GPT-5.4 Mini, with 400K tokens against 262K.
How many benchmarks is the Devstral 2 2512 vs GPT-5.4 Mini comparison based on?
5 benchmarks have a result for both models as of October 22, 2026. Devstral 2 2512's coding index rests on 1 independent benchmark and GPT-5.4 Mini's on 5. Where only one model has a result, the page says so instead of naming a winner.