Model comparison
GLM-4.6 vs Grok 3
GLM-4.6 is the stronger model overall, scoring 41.4 to 39.9 on the Noometry Index.
Last verified . 22 shared benchmarks.
Summary
- They share 22 benchmarks with published results for both. GLM-4.6 scores higher in 6 categories and Grok 3 in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-4.6 leads 23.7 to 13.7.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 47.4% for GLM-4.6 and 61.3% for Grok 3.
- GLM-4.6 has downloadable open weights; the other is API-only.
Side by side
| GLM-4.6 | Grok 3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 41.4 | 39.9 |
| Released | 2025-09-30 | 2025-04-09 |
| Weights | Open | Proprietary |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.60 | — |
| Output $ / M tokens | $2.20 | — |
| Results tracked | 29 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
GLM-4.6: 40.1 (#148), Grok 3: 41.9 (#115)
| Benchmark | GLM-4.6 | Grok 3 |
|---|---|---|
| LMArena Coding | 1449 | 1432 |
| SWE-bench Verified (bash only) | 55.4% | — |
| Aider Polyglot | — | 53.3% |
| LMArena WebDev | 1340 | — |
| SciCode | 38.4% | — |
| WeirdML | — | 37.2% |
| ALE-Bench | 340.82 | — |
Agentic & Tool Use GLM-4.6 leads
GLM-4.6: 32.3 (#66), Grok 3: 30.5 (#76)
| Benchmark | GLM-4.6 | Grok 3 |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| Berkeley Function Calling Leaderboard | 72.4% | — |
| BALROG | — | 29.5% |
Reasoning GLM-4.6 leads
GLM-4.6: 23.7 (#172), Grok 3: 13.7 (#333)
| Benchmark | GLM-4.6 | Grok 3 |
|---|---|---|
| Kagi LLM Benchmark | 47.4% | 61.3% |
| LMArena Hard Prompts | 1440 | 1434 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 36.1% |
| ARC-AGI-1 | — | 5.5% |
| CritPt | 1.1% | — |
| Epoch Capabilities Index | — | 138.33 |
Math GLM-4.6 leads
GLM-4.6: 39.1 (#111), Grok 3: 38.0 (#145)
| Benchmark | GLM-4.6 | Grok 3 |
|---|---|---|
| LMArena Math | 1432 | 1391 |
| FrontierMath (Feb 2025 set) | 3.8% | 3.8% |
| FrontierMath Tier 4 (v1) | 2.1% | 0% |
| OTIS Mock AIME 2024-2025 | — | 55.6% |
| Omni-MATH | — | 46.4% |
| MATH Level 5 | — | 88.7% |
Knowledge Grok 3 leads
GLM-4.6: 40.2 (#124), Grok 3: 46.2 (#82)
| Benchmark | GLM-4.6 | Grok 3 |
|---|---|---|
| Vectara Hallucination Rate | 9.5% | 5.8% |
| LMArena Expert | 1431 | 1421 |
| GPQA Diamond | — | 75.8% |
| MMLU-Pro | — | 78.8% |
| Confabulations | — | 14.2% |
| GPQA (HELM) | — | 65% |
Multilingual GLM-4.6 leads
GLM-4.6: 53.5 (#66), Grok 3: 52.3 (#87)
| Benchmark | GLM-4.6 | Grok 3 |
|---|---|---|
| LMArena Non-English | 1426 | 1410 |
| LMArena Chinese | 1499 | 1448 |
| LMArena French | 1459 | 1460 |
| LMArena German | 1447 | 1431 |
| LMArena Japanese | 1393 | 1387 |
| LMArena Korean | 1400 | 1373 |
| LMArena Russian | 1419 | 1416 |
| LMArena Spanish | 1436 | 1417 |
Instruction Following Too close to call
GLM-4.6: 74.3 (#98), Grok 3: 75.0 (#73)
| Benchmark | GLM-4.6 | Grok 3 |
|---|---|---|
| LMArena Instruction Following | 1410 | 1409 |
| IFEval | — | 88.4% |
Long Context GLM-4.6 leads
GLM-4.6: 43.4 (#94), Grok 3: 38.7 (#192)
| Benchmark | GLM-4.6 | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1422 | 1439 |
| Fiction.LiveBench | — | 58.3% |
Writing & Preference GLM-4.6 leads
GLM-4.6: 61.1 (#90), Grok 3: 55.8 (#141)
| Benchmark | GLM-4.6 | Grok 3 |
|---|---|---|
| LMArena Text | 1440 | 1426 |
| LMArena Creative Writing | 1411 | 1414 |
| EQ-Bench Creative Writing | 1411 | 1186 |
| LMArena Multi-Turn | 1427 | 1425 |
| Short-Story Creative Writing | — | 76.4% |
| WildBench | — | 84.9% |
Frequently asked questions
Is GLM-4.6 better than Grok 3?
GLM-4.6 is the stronger model overall, scoring 41.4 to 39.9 on the Noometry Index.
Is GLM-4.6 or Grok 3 better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 40.1 in the Noometry coding category.
How many benchmarks do GLM-4.6 and Grok 3 share?
22 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Grok 3 has 40.