Model comparison
GLM-5 vs Grok 3
GLM-5 is the stronger model overall, scoring 45.9 to 39.6 on the Noometry Index.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. GLM-5 scores higher in 9 categories and Grok 3 in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5 leads 27.6 to 13.8.
- The biggest single-benchmark swing is ARC-AGI-1: 44.7% for GLM-5 and 5.5% for Grok 3.
- GLM-5 has downloadable open weights; the other is API-only.
Side by side
| GLM-5 | Grok 3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 45.9 | 39.6 |
| Released | 2026-02-11 | 2025-04-09 |
| Weights | Open | Proprietary |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $1 | — |
| Output $ / M tokens | $3.20 | — |
| Results tracked | 45 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
GLM-5: 49.0 (#47), Grok 3: 41.5 (#113)
| Benchmark | GLM-5 | Grok 3 |
|---|---|---|
| WeirdML | 48.2% | 37.2% |
| LMArena Coding | 1461 | 1432 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| Aider Polyglot | — | 53.3% |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| ALE-Bench | 765.62 | — |
Agentic & Tool Use Too close to call
GLM-5: 30.9 (#65), Grok 3: 30.2 (#69)
| Benchmark | GLM-5 | Grok 3 |
|---|---|---|
| Terminal-Bench | 52.4% | — |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| BALROG | — | 29.5% |
| Vending-Bench 2 | 4,432 | — |
Reasoning GLM-5 leads
GLM-5: 27.6 (#100), Grok 3: 13.8 (#322)
| Benchmark | GLM-5 | Grok 3 |
|---|---|---|
| ARC-AGI-2 | 4.9% | 0% |
| SimpleBench | 53.2% | 36.1% |
| Kagi LLM Benchmark | 75% | 61.3% |
| ARC-AGI-1 | 44.7% | 5.5% |
| LMArena Hard Prompts | 1451 | 1433 |
| Epoch Capabilities Index | 145.84 | 138.34 |
| NYT Connections (extended) | 74.8% | — |
| Chess Puzzles | 10% | — |
| ForecastBench | 61 | — |
Math GLM-5 leads
GLM-5: 44.8 (#66), Grok 3: 36.2 (#122)
| Benchmark | GLM-5 | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 55.6% |
| LMArena Math | 1440 | 1390 |
| FrontierMath (Feb 2025 set) | 16.4% | 3.8% |
| FrontierMath Tier 4 (v1) | 2.1% | 0% |
| MathArena Final-Answer Competitions | 65.7% | — |
| Omni-MATH | — | 46.4% |
| MATH Level 5 | — | 88.7% |
Knowledge GLM-5 leads
GLM-5: 52.2 (#59), Grok 3: 46.0 (#76)
| Benchmark | GLM-5 | Grok 3 |
|---|---|---|
| GPQA Diamond | 87.8% | 75.8% |
| Vectara Hallucination Rate | 10.1% | 5.8% |
| LMArena Expert | 1454 | 1420 |
| MMLU-Pro | — | 78.8% |
| Confabulations | — | 14.2% |
| GPQA (HELM) | — | 65% |
Multilingual GLM-5 leads
GLM-5: 53.9 (#57), Grok 3: 52.5 (#81)
| Benchmark | GLM-5 | Grok 3 |
|---|---|---|
| LMArena Non-English | 1430 | 1411 |
| LMArena Chinese | 1511 | 1448 |
| LMArena French | 1455 | 1460 |
| LMArena German | 1445 | 1431 |
| LMArena Japanese | 1416 | 1389 |
| LMArena Korean | 1421 | 1374 |
| LMArena Russian | 1436 | 1416 |
| LMArena Spanish | 1452 | 1417 |
Instruction Following Too close to call
GLM-5: 75.4 (#64), Grok 3: 75.2 (#69)
| Benchmark | GLM-5 | Grok 3 |
|---|---|---|
| LMArena Instruction Following | 1427 | 1409 |
| IFEval | — | 88.4% |
Long Context GLM-5 leads
GLM-5: 44.6 (#58), Grok 3: 38.7 (#187)
| Benchmark | GLM-5 | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1446 | 1439 |
| Fiction.LiveBench | — | 58.3% |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.2 (#38), Grok 3: 56.0 (#136)
| Benchmark | GLM-5 | Grok 3 |
|---|---|---|
| LMArena Text | 1446 | 1426 |
| LMArena Creative Writing | 1438 | 1414 |
| EQ-Bench Creative Writing | 1601 | 1186 |
| LMArena Multi-Turn | 1456 | 1424 |
| Short-Story Creative Writing | — | 76.4% |
| WildBench | — | 84.9% |
Frequently asked questions
Is GLM-5 better than Grok 3?
GLM-5 is the stronger model overall, scoring 45.9 to 39.6 on the Noometry Index.
Is GLM-5 or Grok 3 better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 41.5 in the Noometry coding category.
How many benchmarks do GLM-5 and Grok 3 share?
29 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and Grok 3 has 40.