Model comparison

Devstral Small 2505 vs Grok 4.6

Grok 4.6 is the stronger model overall, scoring 56.9 to 34.3 on the Noometry Index. Devstral Small 2505 costs 20× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.

Last verified October 10, 2026. 2 shared benchmarks.

Devstral Small 2505

Mistral AI

34.3

Rank #233 Reported

Grok 4.6 xAI

56.9

Rank #21 Confirmed

Summary

They share 2 benchmarks with published results for both. Devstral Small 2505 scores higher in 0 categories and Grok 4.6 in 2 categories; 2 gaps are clear of the uncertainty.
The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 19.7.
The biggest single-benchmark swing is SciCode: 28.8% for Devstral Small 2505 and 56.5% for Grok 4.6.
Devstral Small 2505 is cheaper at $0.10 / $0.30 per million input/output tokens, against $2 / $6 for Grok 4.6.
Grok 4.6 accepts more context: 500K tokens versus 128K.
Devstral Small 2505 has downloadable open weights; the other is API-only.

Side by side

Devstral Small 2505 and Grok 4.6 specifications
	Devstral Small 2505	Grok 4.6
Provider	Mistral AI	xAI
Noometry Index	34.3	56.9
Released	2025-05-07	2026-08-12
Weights	Open	Proprietary
Context window	128K	500K
Max output	128K	500K
Input $ / M tokens	$0.10	$2
Output $ / M tokens	$0.30	$6
Results tracked	4	49

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Grok 4.6 leads

Devstral Small 2505: 38.9 (#166), Grok 4.6: 58.5 (#16)

Coding benchmarks
Benchmark	Devstral Small 2505	Grok 4.6
SciCode	28.8%	56.5%
DeepSWE	—	67.5%
FrontierCode	—	48%
SWE-bench Verified (bash only)	56.4%	—
CursorBench	—	41.4%
LMArena WebDev	—	1617
FrontierSWE	—	25.3%
WeirdML	—	67.3%
LMArena Coding	—	1465
ALE-Bench	—	1,508

Agentic & Tool Use Not comparable

Devstral Small 2505: —, Grok 4.6: 39.4 (#27)

Agentic & Tool Use benchmarks
Benchmark	Devstral Small 2505	Grok 4.6
APEX-Agents	—	65.3%
GDP.pdf	—	17.2%
Vending-Bench 2	—	9,047

Reasoning Grok 4.6 leads

Devstral Small 2505: 19.7 (#252), Grok 4.6: 61.4 (#20)

Reasoning benchmarks
Benchmark	Devstral Small 2505	Grok 4.6
CritPt	0%	19.7%
ARC-AGI-2	—	67.1%
SimpleBench	—	75.9%
Kagi LLM Benchmark	37.7%	—
NYT Connections (extended)	—	80%
ARC-AGI-1	—	87.5%
Chess Puzzles	—	40%
EBR-Bench	—	30.5%
LMArena Hard Prompts	—	1447
Mystery Game Puzzles	—	34%
DTBench	—	97.3%
LMCA	—	48.5%
Epoch Capabilities Index	—	156.44

Math Not comparable

Devstral Small 2505: —, Grok 4.6: 67.0 (#24)

Math benchmarks
Benchmark	Devstral Small 2505	Grok 4.6
FrontierMath (Tiers 1-3)	—	66%
FrontierMath Tier 4	—	31.7%
OTIS Mock AIME 2024-2025	—	99.2%
ProofBench	—	51%
LMArena Math	—	1423

Knowledge Not comparable

Devstral Small 2505: —, Grok 4.6: 63.3 (#20)

Knowledge benchmarks
Benchmark	Devstral Small 2505	Grok 4.6
GPQA Diamond	—	94%
SimpleQA Verified	—	49.3%
LMArena Expert	—	1467

Multimodal Not comparable

Devstral Small 2505: —, Grok 4.6: 43.6 (#23)

Multimodal benchmarks
Benchmark	Devstral Small 2505	Grok 4.6
LMArena Vision	—	1263
Blueprint-Bench 2	—	33.2%
Furniture Assembly	—	40%
LMArena Document	—	1452

Multilingual Not comparable

Devstral Small 2505: —, Grok 4.6: 53.0 (#74)

Multilingual benchmarks
Benchmark	Devstral Small 2505	Grok 4.6
LMArena Non-English	—	1420
LMArena Chinese	—	1480
LMArena French	—	1461
LMArena German	—	1431
LMArena Japanese	—	1376
LMArena Korean	—	1397
LMArena Russian	—	1422
LMArena Spanish	—	1404

Instruction Following Not comparable

Devstral Small 2505: —, Grok 4.6: 75.4 (#63)

Instruction Following benchmarks
Benchmark	Devstral Small 2505	Grok 4.6
LMArena Instruction Following	—	1431

Long Context Not comparable

Devstral Small 2505: —, Grok 4.6: 44.5 (#66)

Long Context benchmarks
Benchmark	Devstral Small 2505	Grok 4.6
LMArena Longer Query	—	1454

Writing & Preference Not comparable

Devstral Small 2505: —, Grok 4.6: 62.3 (#80)

Writing & Preference benchmarks
Benchmark	Devstral Small 2505	Grok 4.6
LMArena Text	—	1428
LMArena Creative Writing	—	1428
LMArena Multi-Turn	—	1425

Frequently asked questions

Is Devstral Small 2505 better than Grok 4.6?

Which is cheaper, Devstral Small 2505 or Grok 4.6?

Devstral Small 2505 is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Grok 4.6 lists at $2 and $6.

Is Devstral Small 2505 or Grok 4.6 better for coding?

Grok 4.6 scores higher on coding benchmarks: 58.5 versus 38.9 in the Noometry coding category.

Which has the bigger context window?

Grok 4.6 does, with 500K tokens against 128K.

How many benchmarks do Devstral Small 2505 and Grok 4.6 share?

2 benchmarks have published results for both models. Devstral Small 2505 has 4 scored results on Noometry and Grok 4.6 has 49.