Devstral 2 vs GLM-5.1
Wins 0 of 6 areas
—
Wins 6 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents · Following instructions
GLM-5.1 is the stronger all-rounder.Devstral 2 is cheaper.
Scores updated · 22 tests both models report · How we compare
Where each one wins
Tests won in each of the six areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- CodingWriting and fixing software06GLM-5.16 of 6 tests
- AgentsCarrying out multi-step tasks on its own03GLM-5.13 of 3 tests
- ReasoningHard problems that need careful thinking03GLM-5.13 of 3 tests
- FactsGetting facts right instead of making them up02GLM-5.12 of 2 tests
- Long documentsFinding answers in very long texts01GLM-5.11 of 1 test
- Following instructionsDoing exactly what it is asked01GLM-5.11 of 1 test
Long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
Devstral 2 costs 58% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where GLM-5.1 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+55.4points ahead
- Reasons across sets of long documentsAA-LCR+41.4points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+38.2points ahead
Where Devstral 2 pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 22 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGLM-5.1
- Terminal-Bench 2.1GLM-5.1 by 31.530.361.8+31.5
- LMArena · WebDevGLM-5.1 by 314 rating points11941508+314 rating
- Terminal-Bench HardGLM-5.1 by 24.318.943.2+24.3
- SWE-bench VerifiedGLM-5.1 by 22.453.876.2+22.4
- SciCodeGLM-5.1 by 1232.844.8+12
- SWE-bench MultilingualGLM-5.1 by 1261.373.3+12
AgentsGLM-5.1
- GDPValGLM-5.1 by 28.62.431+28.6
- τ-Bench V3 · BankingGLM-5.1 by 3.110.513.6+3.1
- Terminal-Bench 4.0GLM-5.1 by 202+2
ReasoningGLM-5.1
- GPQA DiamondGLM-5.1 by 27.459.486.8+27.4
- Humanity's Last ExamGLM-5.1 by 26.53.630.1+26.5
- CritPtGLM-5.1 by 4.604.6+4.6
FactsGLM-5.1
- AA-Omniscience · Non-hallucinationGLM-5.1 by 55.414.770.1+55.4
- AA-Omniscience · AccuracyGLM-5.1 by 2.920.823.7+2.9
Other results6 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- τ²-Bench Telecom (AA run)GLM-5.1 by 72.824.997.7+72.8
- AA-OmniscienceGLM-5.1 by 47.6-46.70.8+47.6
- Terminal-Bench 2.0GLM-5.1 by 36.432.669+36.4
- Artificial Analysis Coding IndexGLM-5.1 by 24.531.355.8+24.5
- AA Agentic IndexGLM-5.1 by 20.34.925.2+20.3
- AA IntelligenceGLM-5.1 by 17.58.626.1+17.5
Questions people ask
Which is better, Devstral 2 or GLM-5.1?
GLM-5.1 wins all six areas where both have results: coding, agents, reasoning, facts, long documents and following instructions. Devstral 2 wins none, but costs 58% less.
Which is better for coding?
GLM-5.1. It wins 6 of the 6 coding tests both models report; Devstral 2 wins none.
Which is cheaper?
Devstral 2 costs $0.40 per million input tokens and $2.00 per million output tokens; GLM-5.1 costs $1.38 and $4.40. That makes Devstral 2 about 58% cheaper for the same work.
How do you compare the two?
We use the 22 benchmark tests both models have published scores on. The verdict counts the 16 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 6 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.