Model comparison
GLM-5.2 vs Mistral Large 3
GLM-5.2 is the stronger model overall, scoring 50.9 to 38.6 on the Noometry Index. Mistral Large 3 costs 5.7× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
Mistral AISummary
- They share 21 benchmarks with published results for both. GLM-5.2 scores higher in 8 categories and Mistral Large 3 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.2 leads 42.1 to 14.6.
- The biggest single-benchmark swing is NYT Connections (extended): 74.3% for GLM-5.2 and 7.5% for Mistral Large 3.
- Mistral Large 3 is cheaper at $0.25 / $0.75 per million input/output tokens, against $1.40 / $4.40 for GLM-5.2.
- GLM-5.2 accepts more context: 1M tokens versus 262K.
Side by side
| GLM-5.2 | Mistral Large 3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Mistral AI |
| Noometry Index | 50.9 | 38.6 |
| Released | 2026-06-13 | 2025-12-02 |
| Weights | Open | Open |
| Context window | 1M | 262K |
| Max output | 131K | 8K |
| Input $ / M tokens | $1.40 | $0.25 |
| Output $ / M tokens | $4.40 | $0.75 |
| Results tracked | 50 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
GLM-5.2: 50.9 (#37), Mistral Large 3: 34.1 (#230)
| Benchmark | GLM-5.2 | Mistral Large 3 |
|---|---|---|
| LMArena WebDev | 1604 | 1230 |
| LMArena Coding | 1484 | 1447 |
| SWE-bench Verified | 78.7% | — |
| DeepSWE | 43.8% | — |
| FrontierCode | 24.5% | — |
| SciCode | 50.5% | — |
| WeirdML | 70.1% | — |
| ALE-Bench | 1,047 | — |
Agentic & Tool Use Not comparable
GLM-5.2: 31.6 (#63), Mistral Large 3: —
| Benchmark | GLM-5.2 | Mistral Large 3 |
|---|---|---|
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 31.7% | — |
| GBAEval | 0% | — |
| Vending-Bench 2 | 8,314 | — |
Reasoning GLM-5.2 leads
GLM-5.2: 42.1 (#46), Mistral Large 3: 14.6 (#314)
| Benchmark | GLM-5.2 | Mistral Large 3 |
|---|---|---|
| Kagi LLM Benchmark | 62.6% | 50.9% |
| NYT Connections (extended) | 74.3% | 7.5% |
| LMArena Hard Prompts | 1480 | 1429 |
| ARC-AGI-2 | 22.8% | — |
| SimpleBench | 58.8% | — |
| ARC-AGI-1 | 77% | — |
| CritPt | 20.9% | — |
| Chess Puzzles | 21% | — |
| Thematic Generalization | — | 23% |
| EBR-Bench | 9.5% | — |
| Mystery Game Puzzles | 19% | — |
| DTBench | 93.6% | — |
| LMCA | 45.8% | — |
| Surface Evolver Bench | 55.6% | — |
| Epoch Capabilities Index | 151.76 | — |
Math GLM-5.2 leads
GLM-5.2: 55.4 (#35), Mistral Large 3: 36.0 (#131)
| Benchmark | GLM-5.2 | Mistral Large 3 |
|---|---|---|
| LMArena Math | 1482 | 1415 |
| FrontierMath (Tiers 1-3) | 59.2% | — |
| FrontierMath Tier 4 | 29.3% | — |
| MathArena Final-Answer Competitions | 67.6% | — |
| OTIS Mock AIME 2024-2025 | 86.4% | — |
| ProofBench | 35% | — |
Knowledge GLM-5.2 leads
GLM-5.2: 57.1 (#36), Mistral Large 3: 35.5 (#173)
| Benchmark | GLM-5.2 | Mistral Large 3 |
|---|---|---|
| LMArena Expert | 1484 | 1421 |
| GPQA Diamond | 91.9% | — |
| SimpleQA Verified | 34.2% | — |
| Vectara Hallucination Rate | — | 14.5% |
Multimodal Not comparable
GLM-5.2: —, Mistral Large 3: 37.6 (#65)
| Benchmark | GLM-5.2 | Mistral Large 3 |
|---|---|---|
| LMArena Vision | — | 1222 |
Multilingual GLM-5.2 leads
GLM-5.2: 56.0 (#25), Mistral Large 3: 52.6 (#79)
| Benchmark | GLM-5.2 | Mistral Large 3 |
|---|---|---|
| LMArena Non-English | 1459 | 1412 |
| LMArena Chinese | 1518 | 1448 |
| LMArena French | 1476 | 1456 |
| LMArena German | 1470 | 1438 |
| LMArena Japanese | 1451 | 1399 |
| LMArena Korean | 1445 | 1384 |
| LMArena Russian | 1467 | 1412 |
| LMArena Spanish | 1477 | 1437 |
Instruction Following GLM-5.2 leads
GLM-5.2: 77.1 (#32), Mistral Large 3: 74.1 (#104)
| Benchmark | GLM-5.2 | Mistral Large 3 |
|---|---|---|
| LMArena Instruction Following | 1465 | 1402 |
Long Context GLM-5.2 leads
GLM-5.2: 45.3 (#43), Mistral Large 3: 43.1 (#102)
| Benchmark | GLM-5.2 | Mistral Large 3 |
|---|---|---|
| LMArena Longer Query | 1479 | 1413 |
Writing & Preference GLM-5.2 leads
GLM-5.2: 70.7 (#20), Mistral Large 3: 60.2 (#100)
| Benchmark | GLM-5.2 | Mistral Large 3 |
|---|---|---|
| LMArena Text | 1470 | 1427 |
| LMArena Creative Writing | 1465 | 1385 |
| EQ-Bench Creative Writing | 1757 | 1412 |
| LMArena Multi-Turn | 1468 | 1429 |
| EQ-Bench 4 | 1222 | — |
Frequently asked questions
Is GLM-5.2 better than Mistral Large 3?
GLM-5.2 is the stronger model overall, scoring 50.9 to 38.6 on the Noometry Index. Mistral Large 3 costs 5.7× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Which is cheaper, GLM-5.2 or Mistral Large 3?
Mistral Large 3 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; GLM-5.2 lists at $1.40 and $4.40.
Is GLM-5.2 or Mistral Large 3 better for coding?
GLM-5.2 scores higher on coding benchmarks: 50.9 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
GLM-5.2 does, with 1M tokens against 262K.
How many benchmarks do GLM-5.2 and Mistral Large 3 share?
21 benchmarks have published results for both models. GLM-5.2 has 50 scored results on Noometry and Mistral Large 3 has 24.