Model comparison
GLM-5.1 vs Grok 3
GLM-5.1 is the stronger model overall, scoring 47.8 to 39.9 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. GLM-5.1 scores higher in 8 categories and Grok 3 in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.1 leads 39.1 to 13.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 93.3% for GLM-5.1 and 55.6% for Grok 3.
- GLM-5.1 has downloadable open weights; the other is API-only.
Side by side
| GLM-5.1 | Grok 3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 47.8 | 39.9 |
| Released | 2026-04-07 | 2025-04-09 |
| Weights | Open | Proprietary |
| Context window | 200K | — |
| Max output | 131K | — |
| Input $ / M tokens | $1.40 | — |
| Output $ / M tokens | $4.40 | — |
| Results tracked | 41 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.1 leads
GLM-5.1: 48.7 (#55), Grok 3: 41.9 (#115)
| Benchmark | GLM-5.1 | Grok 3 |
|---|---|---|
| WeirdML | 57.1% | 37.2% |
| LMArena Coding | 1485 | 1432 |
| SWE-bench Verified | 74.2% | — |
| Aider Polyglot | — | 53.3% |
| LMArena WebDev | 1508 | — |
| SciCode | 43.8% | — |
| ALE-Bench | 887.1 | — |
Agentic & Tool Use Grok 3 leads
GLM-5.1: 24.9 (#113), Grok 3: 30.5 (#76)
| Benchmark | GLM-5.1 | Grok 3 |
|---|---|---|
| APEX-Agents | 40.9% | — |
| BALROG | — | 29.5% |
| ExploitBench | 18.1% | — |
| GBAEval | 0% | — |
| Vending-Bench 2 | 5,634 | — |
Reasoning GLM-5.1 leads
GLM-5.1: 39.1 (#60), Grok 3: 13.7 (#333)
| Benchmark | GLM-5.1 | Grok 3 |
|---|---|---|
| SimpleBench | 55.1% | 36.1% |
| LMArena Hard Prompts | 1472 | 1434 |
| Epoch Capabilities Index | 149.84 | 138.33 |
| ARC-AGI-2 | — | 0% |
| Kagi LLM Benchmark | — | 61.3% |
| NYT Connections (extended) | 77.7% | — |
| ARC-AGI-1 | — | 5.5% |
| CritPt | 4.6% | — |
| Chess Puzzles | 19% | — |
| Thematic Generalization | 69.8% | — |
Math GLM-5.1 leads
GLM-5.1: 49.7 (#60), Grok 3: 38.0 (#145)
| Benchmark | GLM-5.1 | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 93.3% | 55.6% |
| LMArena Math | 1473 | 1391 |
| FrontierMath (Feb 2025 set) | 33.4% | 3.8% |
| FrontierMath Tier 4 (v1) | 12.5% | 0% |
| FrontierMath (Tiers 1-3) | 36.8% | — |
| MathArena Final-Answer Competitions | 67.1% | — |
| ProofBench | 22.2% | — |
| Omni-MATH | — | 46.4% |
| MATH Level 5 | — | 88.7% |
Knowledge GLM-5.1 leads
GLM-5.1: 54.9 (#50), Grok 3: 46.2 (#82)
| Benchmark | GLM-5.1 | Grok 3 |
|---|---|---|
| GPQA Diamond | 89.9% | 75.8% |
| LMArena Expert | 1476 | 1421 |
| SimpleQA Verified | 34% | — |
| MMLU-Pro | — | 78.8% |
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 5.8% |
| GPQA (HELM) | — | 65% |
Multilingual GLM-5.1 leads
GLM-5.1: 55.0 (#36), Grok 3: 52.3 (#87)
| Benchmark | GLM-5.1 | Grok 3 |
|---|---|---|
| LMArena Non-English | 1447 | 1410 |
| LMArena Chinese | 1515 | 1448 |
| LMArena French | 1474 | 1460 |
| LMArena German | 1465 | 1431 |
| LMArena Japanese | 1434 | 1387 |
| LMArena Korean | 1418 | 1373 |
| LMArena Russian | 1454 | 1416 |
| LMArena Spanish | 1469 | 1417 |
Instruction Following GLM-5.1 leads
GLM-5.1: 76.3 (#42), Grok 3: 75.0 (#73)
| Benchmark | GLM-5.1 | Grok 3 |
|---|---|---|
| LMArena Instruction Following | 1451 | 1409 |
| IFEval | — | 88.4% |
Long Context GLM-5.1 leads
GLM-5.1: 44.9 (#53), Grok 3: 38.7 (#192)
| Benchmark | GLM-5.1 | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1466 | 1439 |
| Fiction.LiveBench | — | 58.3% |
Writing & Preference GLM-5.1 leads
GLM-5.1: 66.9 (#31), Grok 3: 55.8 (#141)
| Benchmark | GLM-5.1 | Grok 3 |
|---|---|---|
| LMArena Text | 1461 | 1426 |
| LMArena Creative Writing | 1453 | 1414 |
| EQ-Bench Creative Writing | 1592 | 1186 |
| LMArena Multi-Turn | 1472 | 1425 |
| Short-Story Creative Writing | — | 76.4% |
| WildBench | — | 84.9% |
Frequently asked questions
Is GLM-5.1 better than Grok 3?
GLM-5.1 is the stronger model overall, scoring 47.8 to 39.9 on the Noometry Index.
Is GLM-5.1 or Grok 3 better for coding?
GLM-5.1 scores higher on coding benchmarks: 48.7 versus 41.9 in the Noometry coding category.
How many benchmarks do GLM-5.1 and Grok 3 share?
25 benchmarks have published results for both models. GLM-5.1 has 41 scored results on Noometry and Grok 3 has 40.