Model comparison
GLM-5.2 vs Grok 4.20 (Non-Reasoning)
GLM-5.2 is the stronger model overall, scoring 51.1 to 48.6 on the Noometry Index.
Last verified . 37 shared benchmarks.
Summary
- They share 37 benchmarks with published results for both. GLM-5.2 scores higher in 6 categories and Grok 4.20 (Non-Reasoning) in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.3 to 42.3.
- The biggest single-benchmark swing is ARC-AGI-2: 22.8% for GLM-5.2 and 65.1% for Grok 4.20 (Non-Reasoning).
- Grok 4.20 (Non-Reasoning) is cheaper at $1.25 / $2.50 per million input/output tokens, against $1.40 / $4.40 for GLM-5.2.
- GLM-5.2 has downloadable open weights; the other is API-only.
Side by side
| GLM-5.2 | Grok 4.20 (Non-Reasoning) | |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 51.1 | 48.6 |
| Released | 2026-06-13 | 2026-02-17 |
| Weights | Open | Proprietary |
| Context window | 1M | 1M |
| Max output | 131K | 30K |
| Input $ / M tokens | $1.40 | $1.25 |
| Output $ / M tokens | $4.40 | $2.50 |
| Results tracked | 51 | 46 |
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), Grok 4.20 (Non-Reasoning): 42.1 (#112)
| Benchmark | GLM-5.2 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena WebDev | 1603 | 1375 |
| WeirdML | 70.1% | 52.3% |
| LMArena Coding | 1485 | 1459 |
| ALE-Bench | 1,047 | 1,150 |
| SWE-bench Verified | 78.7% | — |
| DeepSWE | 43.8% | — |
| FrontierCode | 24.5% | — |
| SciCode | 50.5% | — |
Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads
GLM-5.2: 32.4 (#63), Grok 4.20 (Non-Reasoning): 34.4 (#46)
| Benchmark | GLM-5.2 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| τ²-bench Banking | 37.1% | 18% |
| Vending-Bench 2 | 8,314 | 4,663 |
| Terminal-Bench | — | 57.3% |
| APEX-Agents | 45.2% | — |
| PostTrainBench | 31.7% | — |
| GBAEval | 0% | — |
| LMArena Search | — | 1189 |
Reasoning Grok 4.20 (Non-Reasoning) leads
GLM-5.2: 42.3 (#52), Grok 4.20 (Non-Reasoning): 52.3 (#32)
| Benchmark | GLM-5.2 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| ARC-AGI-2 | 22.8% | 65.1% |
| Kagi LLM Benchmark | 62.6% | 75% |
| NYT Connections (extended) | 74.3% | 85.4% |
| ARC-AGI-1 | 77% | 89.5% |
| Chess Puzzles | 21% | 24% |
| LMArena Hard Prompts | 1480 | 1451 |
| DTBench | 93.6% | 90.1% |
| LMCA | 45.8% | 38.7% |
| Epoch Capabilities Index | 151.78 | 151.98 |
| SimpleBench | 58.8% | — |
| CritPt | 20.9% | — |
| Thematic Generalization | — | 63.8% |
| EBR-Bench | 9.5% | — |
| Mystery Game Puzzles | 19% | — |
| Surface Evolver Bench | 55.6% | — |
| ForecastBench | — | 61.4 |
Math GLM-5.2 leads
GLM-5.2: 55.7 (#43), Grok 4.20 (Non-Reasoning): 48.2 (#65)
| Benchmark | GLM-5.2 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| FrontierMath (Tiers 1-3) | 59.2% | 44.9% |
| FrontierMath Tier 4 | 29.3% | 17.1% |
| OTIS Mock AIME 2024-2025 | 86.4% | 92.2% |
| ProofBench | 35% | 14% |
| LMArena Math | 1482 | 1455 |
| MathArena Final-Answer Competitions | 67.6% | — |
Knowledge GLM-5.2 leads
GLM-5.2: 57.1 (#40), Grok 4.20 (Non-Reasoning): 52.8 (#60)
| Benchmark | GLM-5.2 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| GPQA Diamond | 91.9% | 89.3% |
| SimpleQA Verified | 34.2% | 30.2% |
| LMArena Expert | 1486 | 1439 |
Multimodal Not comparable
GLM-5.2: —, Grok 4.20 (Non-Reasoning): 33.3 (#98)
| Benchmark | GLM-5.2 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Vision | — | 1263 |
| Blueprint-Bench 2 | — | 0% |
| LMArena Document | — | 1416 |
Multilingual GLM-5.2 leads
GLM-5.2: 55.8 (#26), Grok 4.20 (Non-Reasoning): 54.5 (#40)
| Benchmark | GLM-5.2 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Non-English | 1459 | 1441 |
| LMArena Chinese | 1519 | 1481 |
| LMArena French | 1479 | 1476 |
| LMArena German | 1468 | 1465 |
| LMArena Japanese | 1451 | 1449 |
| LMArena Korean | 1445 | 1417 |
| LMArena Russian | 1466 | 1458 |
| LMArena Spanish | 1477 | 1443 |
Instruction Following GLM-5.2 leads
GLM-5.2: 76.9 (#34), Grok 4.20 (Non-Reasoning): 74.8 (#83)
| Benchmark | GLM-5.2 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Instruction Following | 1465 | 1420 |
Long Context Too close to call
GLM-5.2: 45.3 (#43), Grok 4.20 (Non-Reasoning): 45.5 (#34)
| Benchmark | GLM-5.2 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Longer Query | 1479 | 1437 |
| CL-bench | — | 22.2% |
| CL-bench Life | — | 11.9% |
Writing & Preference GLM-5.2 leads
GLM-5.2: 70.4 (#21), Grok 4.20 (Non-Reasoning): 65.7 (#44)
| Benchmark | GLM-5.2 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Text | 1470 | 1451 |
| LMArena Creative Writing | 1462 | 1438 |
| EQ-Bench Creative Writing | 1757 | 1574 |
| LMArena Multi-Turn | 1469 | 1456 |
| EQ-Bench 4 | 1222 | — |
Frequently asked questions
Is GLM-5.2 better than Grok 4.20 (Non-Reasoning)?
GLM-5.2 is the stronger model overall, scoring 51.1 to 48.6 on the Noometry Index.
Which is cheaper, GLM-5.2 or Grok 4.20 (Non-Reasoning)?
Grok 4.20 (Non-Reasoning) is cheaper. It lists at $1.25 per million input tokens and $2.50 per million output tokens; GLM-5.2 lists at $1.40 and $4.40.
Is GLM-5.2 or Grok 4.20 (Non-Reasoning) better for coding?
GLM-5.2 scores higher on coding benchmarks: 51.3 versus 42.1 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do GLM-5.2 and Grok 4.20 (Non-Reasoning) share?
37 benchmarks have published results for both models. GLM-5.2 has 51 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.