Model comparison
GLM-5.3-Flash vs Mistral 7B
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 23.0 on the Noometry Index.
Last verified . 20 shared benchmarks.
Mistral AISummary
- They share 20 benchmarks with published results for both. GLM-5.3-Flash scores higher in 8 categories and Mistral 7B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-5.3-Flash leads 58.4 to 7.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 93.9% for GLM-5.3-Flash and 0.3% for Mistral 7B.
- GLM-5.3-Flash is cheaper at $0.15 / $0.50 per million input/output tokens, against $0.25 / $0.25 for Mistral 7B.
- GLM-5.3-Flash accepts more context: 1M tokens versus 8K.
Side by side
| GLM-5.3-Flash | Mistral 7B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Mistral AI |
| Noometry Index | 51.8 | 23.0 |
| Released | 2026-08-20 | 2023-09-27 |
| Weights | Open | Open |
| Context window | 1M | 8K |
| Max output | 131K | 8K |
| Input $ / M tokens | $0.15 | $0.25 |
| Output $ / M tokens | $0.50 | $0.25 |
| Results tracked | 40 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3-Flash leads
GLM-5.3-Flash: 53.1 (#31), Mistral 7B: 26.4 (#326)
| Benchmark | GLM-5.3-Flash | Mistral 7B |
|---|---|---|
| LMArena Coding | 1508 | 1082 |
| DeepSWE | 63.4% | — |
| FrontierCode | 31.8% | — |
| CursorBench | 36.8% | — |
| LMArena WebDev | 1609 | — |
| FrontierSWE | 18.1% | — |
| SciCode | 51.6% | — |
| BigCodeBench Instruct | — | 19.5% |
| BigCodeBench Complete | — | 27.3% |
| ALE-Bench | 303.55 | — |
| HumanEval+ | — | 36% |
| MBPP+ | — | 42.1% |
Agentic & Tool Use Not comparable
GLM-5.3-Flash: 34.2 (#47), Mistral 7B: —
| Benchmark | GLM-5.3-Flash | Mistral 7B |
|---|---|---|
| APEX-Agents | 52.8% | — |
| GDP.pdf | 14% | — |
Reasoning GLM-5.3-Flash leads
GLM-5.3-Flash: 48.0 (#42), Mistral 7B: 13.1 (#336)
| Benchmark | GLM-5.3-Flash | Mistral 7B |
|---|---|---|
| Chess Puzzles | 14% | 0% |
| LMArena Hard Prompts | 1491 | 1067 |
| Epoch Capabilities Index | 151.88 | 112.21 |
| ARC-AGI-2 | 65.8% | — |
| ARC-AGI-1 | 91% | — |
| CritPt | 15.4% | — |
| Mystery Game Puzzles | 8% | — |
| DTBench | — | 42.5% |
| Surface Evolver Bench | 52.5% | — |
| Adversarial NLI | — | 47.1% |
| Bench to the Future 3 | 0.15 | — |
| BIG-Bench Hard | — | 56.1% |
| HellaSwag | — | 81% |
| PIQA | — | 83% |
| WinoGrande | — | 75.3% |
Math GLM-5.3-Flash leads
GLM-5.3-Flash: 53.3 (#47), Mistral 7B: 8.1 (#325)
| Benchmark | GLM-5.3-Flash | Mistral 7B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 93.9% | 0.3% |
| LMArena Math | 1500 | 1085 |
| FrontierMath (Tiers 1-3) | 55.8% | — |
| FrontierMath Tier 4 | 17.1% | — |
| ProofBench | 21% | — |
| MATH Level 5 | — | 3.7% |
| GSM8K | — | 54.4% |
Knowledge GLM-5.3-Flash leads
GLM-5.3-Flash: 58.4 (#36), Mistral 7B: 7.4 (#311)
| Benchmark | GLM-5.3-Flash | Mistral 7B |
|---|---|---|
| GPQA Diamond | 90.2% | 15.2% |
| LMArena Expert | 1513 | 1036 |
| ARC (AI2) Challenge | — | 78.6% |
| BoolQ | — | 87.4% |
| MMLU | — | 62.5% |
| OpenBookQA | — | 79.8% |
| TriviaQA | — | 75.2% |
Multimodal Not comparable
GLM-5.3-Flash: 42.8 (#27), Mistral 7B: —
| Benchmark | GLM-5.3-Flash | Mistral 7B |
|---|---|---|
| LMArena Vision | 1296 | — |
Multilingual GLM-5.3-Flash leads
GLM-5.3-Flash: 56.0 (#25), Mistral 7B: 25.8 (#283)
| Benchmark | GLM-5.3-Flash | Mistral 7B |
|---|---|---|
| LMArena Non-English | 1462 | 1012 |
| LMArena Chinese | 1527 | 1009 |
| LMArena French | 1496 | 1037 |
| LMArena German | 1470 | 987 |
| LMArena Japanese | 1429 | 878 |
| LMArena Russian | 1469 | 1018 |
| LMArena Spanish | 1471 | 1026 |
| LMArena Korean | 1446 | — |
Instruction Following GLM-5.3-Flash leads
GLM-5.3-Flash: 77.5 (#20), Mistral 7B: 54.2 (#280)
| Benchmark | GLM-5.3-Flash | Mistral 7B |
|---|---|---|
| LMArena Instruction Following | 1478 | 1060 |
Long Context GLM-5.3-Flash leads
GLM-5.3-Flash: 45.4 (#39), Mistral 7B: 32.2 (#271)
| Benchmark | GLM-5.3-Flash | Mistral 7B |
|---|---|---|
| LMArena Longer Query | 1482 | 1060 |
Writing & Preference GLM-5.3-Flash leads
GLM-5.3-Flash: 65.3 (#50), Mistral 7B: 30.7 (#286)
| Benchmark | GLM-5.3-Flash | Mistral 7B |
|---|---|---|
| LMArena Text | 1471 | 1090 |
| LMArena Creative Writing | 1442 | 1068 |
| LMArena Multi-Turn | 1467 | 1062 |
Frequently asked questions
Is GLM-5.3-Flash better than Mistral 7B?
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 23.0 on the Noometry Index.
Which is cheaper, GLM-5.3-Flash or Mistral 7B?
GLM-5.3-Flash is cheaper. It lists at $0.15 per million input tokens and $0.50 per million output tokens; Mistral 7B lists at $0.25 and $0.25.
Is GLM-5.3-Flash or Mistral 7B better for coding?
GLM-5.3-Flash scores higher on coding benchmarks: 53.1 versus 26.4 in the Noometry coding category.
Which has the bigger context window?
GLM-5.3-Flash does, with 1M tokens against 8K.
How many benchmarks do GLM-5.3-Flash and Mistral 7B share?
20 benchmarks have published results for both models. GLM-5.3-Flash has 40 scored results on Noometry and Mistral 7B has 37.