Enter opens the highlighted result. Esc closes.

Devstral 2 2512 vs Solar Pro 4

Solar Pro 4 leads Devstral 2 2512 on the Artificial Analysis Intelligence Index, 28.2 to 8.6, and Solar Pro 4 costs 3.3x less per output token ($0.60 vs $2.00 per million). Updated October 22, 2026; 2 benchmarks have a result for both.

Devstral 2 2512Mistral AI
Coding
No data
Overall
No data
Output $/M
$2.00
0
Solar Pro 4Upstage
Coding
No data
Overall
No data
Output $/M
$0.60
2
vs boards won of 2 shared

Devstral 2 2512 or Solar Pro 4: which to pick

Neither model has a coding score yet, so the two cannot be ranked on coding.

Pick Devstral 2 2512 for work measured by AA Coding Index or AA Intelligence Index, where it leads.

Shared results
2
Devstral 2 2512 only
4
Solar Pro 4 only
43
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.

At a glance

Coding index
Devstral 2 2512 not benchmarked, Solar Pro 4 not benchmarked.
Widest category
Coding, 21.4 points apart on average over 1 shared benchmark.
Widest single gap
AA Coding Index: Devstral 2 2512 31.3 against Solar Pro 4 52.7, 21.4 points apart on a 0 to 100 scale.
Output price
Devstral 2 2512 $2.00 against Solar Pro 4 $0.60 per million output tokens, 3.3x apart.
Context window
Devstral 2 2512 takes 262K tokens, Solar Pro 4 524K.

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.

Boards won, where both have a result
CategoryDevstral 2 2512BarsSolar Pro 4Ties
Coding01
Intelligence01
All shared boards02
CodingLike-for-likeSix of these boards feed the Vibe Coding Index01
Coding benchmarks for Devstral 2 2512 and Solar Pro 4
BenchmarkDevstral 2 2512Solar Pro 4Leader
SWE-bench Verified53.8%Not measuredOnly Devstral 2 2512 measured
LMArena WebDevNot measured1371Only Solar Pro 4 measured
AA Coding Index31.352.7Solar Pro 4
LiveCodeBench (Vals AI)51.8%Not measuredOnly Devstral 2 2512 measured
SWE-bench Verified (Vals AI)62.8%Not measuredOnly Devstral 2 2512 measured
SciCodeNot measured48.0%Only Solar Pro 4 measured
ALE-BenchNot measured179.3Only Solar Pro 4 measured
LMArena Text: codingNot measured1438Only Solar Pro 4 measured
LMArena Text: software and ITNot measured1431Only Solar Pro 4 measured
LMArena WebDev: ReactNot measured1363Only Solar Pro 4 measured
LMArena WebDev: HTMLNot measured1371Only Solar Pro 4 measured
Neither model has a result on the other 28 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Devstral 2 2512 and Solar Pro 4
BenchmarkDevstral 2 2512Solar Pro 4Leader
AA Intelligence Index8.628.2Solar Pro 4
Neither model has a result on the other 1 board here
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
Agentic work benchmarks for Devstral 2 2512 and Solar Pro 4
BenchmarkDevstral 2 2512Solar Pro 4Leader
AA Agentic Index2.3Not measuredOnly Devstral 2 2512 measured
Design Arena: agents building web appsNot measured1105Only Solar Pro 4 measured
Neither model has a result on the other 39 boards here
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
Reasoning benchmarks for Devstral 2 2512 and Solar Pro 4
BenchmarkDevstral 2 2512Solar Pro 4Leader
GPQA Diamond (Artificial Analysis run)Not measured89.1%Only Solar Pro 4 measured
Humanity's Last Exam (Artificial Analysis run)Not measured29.8%Only Solar Pro 4 measured
LMArena Text: hard promptsNot measured1400Only Solar Pro 4 measured
LMArena Text: expertNot measured1428Only Solar Pro 4 measured
LMArena Text: hard prompts in EnglishNot measured1427Only Solar Pro 4 measured
Neither model has a result on the other 17 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
Math benchmarks for Devstral 2 2512 and Solar Pro 4
BenchmarkDevstral 2 2512Solar Pro 4Leader
LMArena Text: mathNot measured1421Only Solar Pro 4 measured
LMArena Text: mathematical workNot measured1427Only Solar Pro 4 measured
Neither model has a result on the other 11 boards here
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
Knowledge benchmarks for Devstral 2 2512 and Solar Pro 4
BenchmarkDevstral 2 2512Solar Pro 4Leader
LMArena Text: scienceNot measured1404Only Solar Pro 4 measured
LMArena Text: medicine and healthcareNot measured1374Only Solar Pro 4 measured
LMArena Text: legal and governmentNot measured1383Only Solar Pro 4 measured
LMArena Text: business and financeNot measured1381Only Solar Pro 4 measured
Neither model has a result on the other 14 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
Web design benchmarks for Devstral 2 2512 and Solar Pro 4
BenchmarkDevstral 2 2512Solar Pro 4Leader
Design Arena: websitesNot measured1176Only Solar Pro 4 measured
Design Arena: UI componentsNot measured1147Only Solar Pro 4 measured
Design Arena: game devNot measured1169Only Solar Pro 4 measured
Design Arena: data vizNot measured1178Only Solar Pro 4 measured
Design Arena: 3DNot measured1178Only Solar Pro 4 measured
Design Arena: all code categoriesNot measured1181Only Solar Pro 4 measured
Neither model has a result on the other 3 boards here
GeneralDirectional onlyPublic boards, shown as published, kept out of the index18 boards, none shared
General benchmarks for Devstral 2 2512 and Solar Pro 4
BenchmarkDevstral 2 2512Solar Pro 4Leader
Long context reasoning (AA-LCR)Not measured70.7%Only Solar Pro 4 measured
LMArena TextNot measured1388Only Solar Pro 4 measured
LMArena Text: instruction followingNot measured1375Only Solar Pro 4 measured
LMArena Text: creative writingNot measured1317Only Solar Pro 4 measured
LMArena Text: EnglishNot measured1414Only Solar Pro 4 measured
LMArena Text: non-EnglishNot measured1363Only Solar Pro 4 measured
LMArena Text: ChineseNot measured1422Only Solar Pro 4 measured
LMArena Text: GermanNot measured1358Only Solar Pro 4 measured
LMArena Text: JapaneseNot measured1320Only Solar Pro 4 measured
LMArena Text: KoreanNot measured1383Only Solar Pro 4 measured
LMArena Text: FrenchNot measured1400Only Solar Pro 4 measured
LMArena Text: SpanishNot measured1409Only Solar Pro 4 measured
LMArena Text: RussianNot measured1357Only Solar Pro 4 measured
LMArena Text: PolishNot measured1304Only Solar Pro 4 measured
LMArena Text: multi-turnNot measured1385Only Solar Pro 4 measured
LMArena Text: longer queriesNot measured1382Only Solar Pro 4 measured
LMArena Text: writing and languageNot measured1344Only Solar Pro 4 measured
LMArena Text: entertainment and mediaNot measured1332Only Solar Pro 4 measured
Neither model has a result on the other 28 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Devstral 2 2512 and Solar Pro 4
BenchmarkDevstral 2 2512Solar Pro 4Leader
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
Solar Pro 4$0.0004Fits in one request

Solar Pro 4 costs less

Repository review

50K input + 3K output tokens

Devstral 2 2512$0.026Fits in one request
Solar Pro 4$0.0093Fits in one request

Solar Pro 4 costs less

Agent loop

200K input + 10K output tokens

Devstral 2 2512$0.100Fits in one request
Solar Pro 4$0.036Fits in one request

Solar Pro 4 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.

Specifications of Devstral 2 2512 and Solar Pro 4
SpecificationMistral Devstral 2 2512Upstage Solar Pro 4
MakerMistral AIUpstage
WeightsOpen weightsNo public weights listed
ReleasedDecember 9, 2025August 10, 2026
Context window262K524K
Input $/Mtok$0.40$0.15
Output $/Mtok$2.00$0.60
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it11
Cost per solved task$1.19Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Devstral 2 2512 better than Solar Pro 4 for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Devstral 2 2512 for work measured by AA Coding Index or AA Intelligence Index, where it leads.

Which is cheaper, Devstral 2 2512 or Solar Pro 4?

Devstral 2 2512 costs $0.40 per million input tokens and $2.00 per million output tokens; Solar Pro 4 costs $0.15 and $0.60. Cost per solved task is known for Devstral 2 2512 only.

Which has the bigger context window, Devstral 2 2512 or Solar Pro 4?

Solar Pro 4, with 524K tokens against 262K.

How many benchmarks is the Devstral 2 2512 vs Solar Pro 4 comparison based on?

2 benchmarks have a result for both models as of October 22, 2026. Devstral 2 2512's coding index rests on 1 independent benchmark and Solar Pro 4's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next