Model comparison
GLM-5.2 vs Mixtral 8x22B
GLM-5.2 is the stronger model overall, scoring 51.1 to 27.1 on the Noometry Index.
Last verified . 21 shared benchmarks.
Mistral AISummary
- They share 21 benchmarks with published results for both. GLM-5.2 scores higher in 9 categories and Mixtral 8x22B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-5.2 leads 57.1 to 15.1.
- The biggest single-benchmark swing is WeirdML: 70.1% for GLM-5.2 and 3.2% for Mixtral 8x22B.
- GLM-5.2 is cheaper at $1.40 / $4.40 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- GLM-5.2 accepts more context: 1M tokens versus 64K.
Side by side
| GLM-5.2 | Mixtral 8x22B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Mistral AI |
| Noometry Index | 51.1 | 27.1 |
| Released | 2026-06-13 | 2024-04-17 |
| Weights | Open | Open |
| Context window | 1M | 64K |
| Max output | 131K | 64K |
| Input $ / M tokens | $1.40 | $2 |
| Output $ / M tokens | $4.40 | $6 |
| Results tracked | 51 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
GLM-5.2: 51.3 (#41), Mixtral 8x22B: 24.2 (#329)
| Benchmark | GLM-5.2 | Mixtral 8x22B |
|---|---|---|
| WeirdML | 70.1% | 3.2% |
| LMArena Coding | 1485 | 1166 |
| SWE-bench Verified | 78.7% | — |
| DeepSWE | 43.8% | — |
| FrontierCode | 24.5% | — |
| LMArena WebDev | 1603 | — |
| SciCode | 50.5% | — |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
| ALE-Bench | 1,047 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |
Agentic & Tool Use GLM-5.2 leads
GLM-5.2: 32.4 (#63), Mixtral 8x22B: 23.1 (#127)
| Benchmark | GLM-5.2 | Mixtral 8x22B |
|---|---|---|
| APEX-Agents | 45.2% | — |
| τ²-bench Banking | 37.1% | — |
| Cybench | — | 7.5% |
| PostTrainBench | 31.7% | — |
| GBAEval | 0% | — |
| Vending-Bench 2 | 8,314 | — |
Reasoning GLM-5.2 leads
GLM-5.2: 42.3 (#52), Mixtral 8x22B: 19.9 (#248)
| Benchmark | GLM-5.2 | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1480 | 1150 |
| DTBench | 93.6% | 55.1% |
| Epoch Capabilities Index | 151.78 | 122.03 |
| ARC-AGI-2 | 22.8% | — |
| SimpleBench | 58.8% | — |
| Kagi LLM Benchmark | 62.6% | — |
| NYT Connections (extended) | 74.3% | — |
| ARC-AGI-1 | 77% | — |
| CritPt | 20.9% | — |
| Chess Puzzles | 21% | — |
| EBR-Bench | 9.5% | — |
| Mystery Game Puzzles | 19% | — |
| LMCA | 45.8% | — |
| Surface Evolver Bench | 55.6% | — |
| ForecastBench | — | 56.3 |
Math GLM-5.2 leads
GLM-5.2: 55.7 (#43), Mixtral 8x22B: 22.9 (#275)
| Benchmark | GLM-5.2 | Mixtral 8x22B |
|---|---|---|
| LMArena Math | 1482 | 1184 |
| 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% | — |
| Omni-MATH | — | 16.3% |
| MATH Level 5 | — | 24.2% |
Knowledge GLM-5.2 leads
GLM-5.2: 57.1 (#40), Mixtral 8x22B: 15.1 (#293)
| Benchmark | GLM-5.2 | Mixtral 8x22B |
|---|---|---|
| GPQA Diamond | 91.9% | 34.1% |
| LMArena Expert | 1486 | 1113 |
| SimpleQA Verified | 34.2% | — |
| MMLU-Pro | — | 46% |
| GPQA (HELM) | — | 33.4% |
| MMLU | — | 77.8% |
Multilingual GLM-5.2 leads
GLM-5.2: 55.8 (#26), Mixtral 8x22B: 32.8 (#255)
| Benchmark | GLM-5.2 | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1459 | 1128 |
| LMArena Chinese | 1519 | 1116 |
| LMArena French | 1479 | 1166 |
| LMArena German | 1468 | 1141 |
| LMArena Japanese | 1451 | 1037 |
| LMArena Korean | 1445 | 1057 |
| LMArena Russian | 1466 | 1158 |
| LMArena Spanish | 1477 | 1151 |
Instruction Following GLM-5.2 leads
GLM-5.2: 76.9 (#34), Mixtral 8x22B: 57.7 (#266)
| Benchmark | GLM-5.2 | Mixtral 8x22B |
|---|---|---|
| LMArena Instruction Following | 1465 | 1147 |
| IFEval | — | 72.4% |
Long Context GLM-5.2 leads
GLM-5.2: 45.3 (#43), Mixtral 8x22B: 34.7 (#247)
| Benchmark | GLM-5.2 | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1479 | 1144 |
Writing & Preference GLM-5.2 leads
GLM-5.2: 70.4 (#21), Mixtral 8x22B: 36.9 (#262)
| Benchmark | GLM-5.2 | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1470 | 1162 |
| LMArena Creative Writing | 1462 | 1141 |
| LMArena Multi-Turn | 1469 | 1130 |
| EQ-Bench Creative Writing | 1757 | — |
| WildBench | — | 71.1% |
| EQ-Bench 4 | 1222 | — |
Frequently asked questions
Is GLM-5.2 better than Mixtral 8x22B?
GLM-5.2 is the stronger model overall, scoring 51.1 to 27.1 on the Noometry Index.
Which is cheaper, GLM-5.2 or Mixtral 8x22B?
GLM-5.2 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is GLM-5.2 or Mixtral 8x22B better for coding?
GLM-5.2 scores higher on coding benchmarks: 51.3 versus 24.2 in the Noometry coding category.
Which has the bigger context window?
GLM-5.2 does, with 1M tokens against 64K.
How many benchmarks do GLM-5.2 and Mixtral 8x22B share?
21 benchmarks have published results for both models. GLM-5.2 has 51 scored results on Noometry and Mixtral 8x22B has 34.