Model comparison
GLM-5.3 vs Grok 3
GLM-5.3 is the stronger model overall, scoring 54.8 to 39.9 on the Noometry Index.
Last verified . 22 shared benchmarks.
Summary
- They share 22 benchmarks with published results for both. GLM-5.3 scores higher in 9 categories and Grok 3 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.3 leads 46.1 to 13.7.
- The biggest single-benchmark swing is WeirdML: 75.4% for GLM-5.3 and 37.2% for Grok 3.
- GLM-5.3 has downloadable open weights; the other is API-only.
Side by side
| GLM-5.3 | Grok 3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 54.8 | 39.9 |
| Released | 2026-08-14 | 2025-04-09 |
| Weights | Open | Proprietary |
| Context window | 1M | — |
| Max output | 131K | — |
| Input $ / M tokens | $1.40 | — |
| Output $ / M tokens | $4.40 | — |
| Results tracked | 42 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3 leads
GLM-5.3: 59.5 (#14), Grok 3: 41.9 (#115)
| Benchmark | GLM-5.3 | Grok 3 |
|---|---|---|
| WeirdML | 75.4% | 37.2% |
| LMArena Coding | 1496 | 1432 |
| DeepSWE | 69% | — |
| FrontierCode | 40.1% | — |
| Aider Polyglot | — | 53.3% |
| CursorBench | 42.6% | — |
| LMArena WebDev | 1622 | — |
| FrontierSWE | 30.2% | — |
| SciCode | 59% | — |
| ALE-Bench | 1,317 | — |
Agentic & Tool Use GLM-5.3 leads
GLM-5.3: 36.4 (#38), Grok 3: 30.5 (#76)
| Benchmark | GLM-5.3 | Grok 3 |
|---|---|---|
| APEX-Agents | 56.6% | — |
| BALROG | — | 29.5% |
| Vending-Bench 2 | 8,164 | — |
Reasoning GLM-5.3 leads
GLM-5.3: 46.1 (#46), Grok 3: 13.7 (#333)
| Benchmark | GLM-5.3 | Grok 3 |
|---|---|---|
| LMArena Hard Prompts | 1489 | 1434 |
| Epoch Capabilities Index | 155.61 | 138.33 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 36.1% |
| Kagi LLM Benchmark | — | 61.3% |
| NYT Connections (extended) | 74.2% | — |
| ARC-AGI-1 | — | 5.5% |
| CritPt | 19.1% | — |
| Chess Puzzles | 21% | — |
| Mystery Game Puzzles | 33% | — |
| DTBench | 87.7% | — |
| LMCA | 55.5% | — |
| Bench to the Future 3 | 0.15 | — |
Math GLM-5.3 leads
GLM-5.3: 62.3 (#33), Grok 3: 38.0 (#145)
| Benchmark | GLM-5.3 | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.1% | 55.6% |
| LMArena Math | 1489 | 1391 |
| FrontierMath (Tiers 1-3) | 68.8% | — |
| FrontierMath Tier 4 | 29.3% | — |
| ProofBench | 49% | — |
| Omni-MATH | — | 46.4% |
| MATH Level 5 | — | 88.7% |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge GLM-5.3 leads
GLM-5.3: 58.3 (#37), Grok 3: 46.2 (#82)
| Benchmark | GLM-5.3 | Grok 3 |
|---|---|---|
| GPQA Diamond | 90.9% | 75.8% |
| LMArena Expert | 1516 | 1421 |
| SimpleQA Verified | 41% | — |
| MMLU-Pro | — | 78.8% |
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 5.8% |
| GPQA (HELM) | — | 65% |
Multilingual GLM-5.3 leads
GLM-5.3: 55.7 (#28), Grok 3: 52.3 (#87)
| Benchmark | GLM-5.3 | Grok 3 |
|---|---|---|
| LMArena Non-English | 1457 | 1410 |
| LMArena Chinese | 1528 | 1448 |
| LMArena French | 1499 | 1460 |
| LMArena German | 1499 | 1431 |
| LMArena Japanese | 1453 | 1387 |
| LMArena Korean | 1472 | 1373 |
| LMArena Russian | 1463 | 1416 |
| LMArena Spanish | 1460 | 1417 |
Instruction Following GLM-5.3 leads
GLM-5.3: 77.5 (#23), Grok 3: 75.0 (#73)
| Benchmark | GLM-5.3 | Grok 3 |
|---|---|---|
| LMArena Instruction Following | 1477 | 1409 |
| IFEval | — | 88.4% |
Long Context GLM-5.3 leads
GLM-5.3: 45.4 (#41), Grok 3: 38.7 (#192)
| Benchmark | GLM-5.3 | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1482 | 1439 |
| Fiction.LiveBench | — | 58.3% |
Writing & Preference GLM-5.3 leads
GLM-5.3: 75.7 (#6), Grok 3: 55.8 (#141)
| Benchmark | GLM-5.3 | Grok 3 |
|---|---|---|
| LMArena Text | 1471 | 1426 |
| LMArena Creative Writing | 1457 | 1414 |
| EQ-Bench Creative Writing | 2075 | 1186 |
| LMArena Multi-Turn | 1472 | 1425 |
| Short-Story Creative Writing | — | 76.4% |
| WildBench | — | 84.9% |
Frequently asked questions
Is GLM-5.3 better than Grok 3?
GLM-5.3 is the stronger model overall, scoring 54.8 to 39.9 on the Noometry Index.
Is GLM-5.3 or Grok 3 better for coding?
GLM-5.3 scores higher on coding benchmarks: 59.5 versus 41.9 in the Noometry coding category.
How many benchmarks do GLM-5.3 and Grok 3 share?
22 benchmarks have published results for both models. GLM-5.3 has 42 scored results on Noometry and Grok 3 has 40.