Model comparison
GLM-5.3 vs Mistral Large
GLM-5.3 is the stronger model overall, scoring 54.8 to 31.9 on the Noometry Index.
Last verified . 26 shared benchmarks.
Mistral AISummary
- They share 26 benchmarks with published results for both. GLM-5.3 scores higher in 9 categories and Mistral Large in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.3 leads 62.3 to 18.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 91.1% for GLM-5.3 and 8.5% for Mistral Large.
- GLM-5.3 is cheaper at $1.40 / $4.40 per million input/output tokens, against $2 / $6 for Mistral Large.
- GLM-5.3 accepts more context: 1M tokens versus 131K.
Side by side
| GLM-5.3 | Mistral Large | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Mistral AI |
| Noometry Index | 54.8 | 31.9 |
| Released | 2026-08-14 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 1M | 131K |
| Max output | 131K | 16K |
| Input $ / M tokens | $1.40 | $2 |
| Output $ / M tokens | $4.40 | $6 |
| Results tracked | 42 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3 leads
GLM-5.3: 59.5 (#14), Mistral Large: 34.3 (#240)
| Benchmark | GLM-5.3 | Mistral Large |
|---|---|---|
| SciCode | 59% | 36.2% |
| LMArena Coding | 1496 | 1277 |
| ALE-Bench | 1,317 | 264.7 |
| DeepSWE | 69% | — |
| FrontierCode | 40.1% | — |
| CursorBench | 42.6% | — |
| LMArena WebDev | 1622 | — |
| FrontierSWE | 30.2% | — |
| WeirdML | 75.4% | — |
| BigCodeBench Instruct | — | 30% |
| LiveBench Coding | — | 47.1% |
| BigCodeBench Complete | — | 38.3% |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use GLM-5.3 leads
GLM-5.3: 36.4 (#38), Mistral Large: 28.6 (#89)
| Benchmark | GLM-5.3 | Mistral Large |
|---|---|---|
| APEX-Agents | 56.6% | — |
| Berkeley Function Calling Leaderboard | — | 38.4% |
| Vending-Bench 2 | 8,164 | — |
Reasoning GLM-5.3 leads
GLM-5.3: 46.1 (#46), Mistral Large: 15.8 (#310)
| Benchmark | GLM-5.3 | Mistral Large |
|---|---|---|
| CritPt | 19.1% | 0% |
| LMArena Hard Prompts | 1489 | 1257 |
| DTBench | 87.7% | 65.1% |
| LMCA | 55.5% | 16.7% |
| Epoch Capabilities Index | 155.61 | 128.52 |
| SimpleBench | — | 22.5% |
| NYT Connections (extended) | 74.2% | — |
| Chess Puzzles | 21% | — |
| LiveBench Reasoning | — | 43.5% |
| Mystery Game Puzzles | 33% | — |
| LiveBench Data Analysis | — | 50.1% |
| Bench to the Future 3 | 0.15 | — |
| ForecastBench | — | 57.1 |
| LiveBench | — | 48.4% |
Math GLM-5.3 leads
GLM-5.3: 62.3 (#33), Mistral Large: 18.2 (#291)
| Benchmark | GLM-5.3 | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.1% | 8.5% |
| LMArena Math | 1489 | 1262 |
| FrontierMath (Tiers 1-3) | 68.8% | — |
| FrontierMath Tier 4 | 29.3% | — |
| ProofBench | 49% | — |
| Omni-MATH | — | 28.1% |
| LiveBench Math | — | 42.5% |
| MATH Level 5 | — | 50.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge GLM-5.3 leads
GLM-5.3: 58.3 (#37), Mistral Large: 30.1 (#230)
| Benchmark | GLM-5.3 | Mistral Large |
|---|---|---|
| GPQA Diamond | 90.9% | 51.3% |
| LMArena Expert | 1516 | 1232 |
| SimpleQA Verified | 41% | — |
| MMLU-Pro | — | 59.9% |
| Confabulations | — | 21.4% |
| Vectara Hallucination Rate | — | 4.5% |
| GPQA (HELM) | — | 43.5% |
| MMLU | — | 80% |
Multilingual GLM-5.3 leads
GLM-5.3: 55.7 (#28), Mistral Large: 40.0 (#219)
| Benchmark | GLM-5.3 | Mistral Large |
|---|---|---|
| LMArena Non-English | 1457 | 1237 |
| LMArena Chinese | 1528 | 1240 |
| LMArena French | 1499 | 1325 |
| LMArena German | 1499 | 1254 |
| LMArena Japanese | 1453 | 1188 |
| LMArena Korean | 1472 | 1202 |
| LMArena Russian | 1463 | 1257 |
| LMArena Spanish | 1460 | 1268 |
Instruction Following GLM-5.3 leads
GLM-5.3: 77.5 (#23), Mistral Large: 67.9 (#191)
| Benchmark | GLM-5.3 | Mistral Large |
|---|---|---|
| LMArena Instruction Following | 1477 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
| IFEval | — | 87.7% |
Long Context GLM-5.3 leads
GLM-5.3: 45.4 (#41), Mistral Large: 38.3 (#199)
| Benchmark | GLM-5.3 | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1482 | 1261 |
Writing & Preference GLM-5.3 leads
GLM-5.3: 75.7 (#6), Mistral Large: 40.7 (#242)
| Benchmark | GLM-5.3 | Mistral Large |
|---|---|---|
| LMArena Text | 1471 | 1266 |
| LMArena Creative Writing | 1457 | 1243 |
| EQ-Bench Creative Writing | 2075 | 985 |
| LMArena Multi-Turn | 1472 | 1260 |
| Short-Story Creative Writing | — | 69% |
| WildBench | — | 80.1% |
| LiveBench Language | — | 39.4% |
Frequently asked questions
Is GLM-5.3 better than Mistral Large?
GLM-5.3 is the stronger model overall, scoring 54.8 to 31.9 on the Noometry Index.
Which is cheaper, GLM-5.3 or Mistral Large?
GLM-5.3 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; Mistral Large lists at $2 and $6.
Is GLM-5.3 or Mistral Large better for coding?
GLM-5.3 scores higher on coding benchmarks: 59.5 versus 34.3 in the Noometry coding category.
Which has the bigger context window?
GLM-5.3 does, with 1M tokens against 131K.
How many benchmarks do GLM-5.3 and Mistral Large share?
26 benchmarks have published results for both models. GLM-5.3 has 42 scored results on Noometry and Mistral Large has 51.