Model comparison
GLM-5.1 vs Grok 4.6
Grok 4.6 is the stronger model overall, scoring 56.9 to 47.8 on the Noometry Index.
Last verified . 33 shared benchmarks.
Summary
- They share 33 benchmarks with published results for both. GLM-5.1 scores higher in 4 categories and Grok 4.6 in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 39.1.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 36.8% for GLM-5.1 and 66% for Grok 4.6.
- GLM-5.1 is cheaper at $1.40 / $4.40 per million input/output tokens, against $2 / $6 for Grok 4.6.
- Grok 4.6 accepts more context: 500K tokens versus 200K.
- GLM-5.1 has downloadable open weights; the other is API-only.
Side by side
| GLM-5.1 | Grok 4.6 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 47.8 | 56.9 |
| Released | 2026-04-07 | 2026-08-12 |
| Weights | Open | Proprietary |
| Context window | 200K | 500K |
| Max output | 131K | 500K |
| Input $ / M tokens | $1.40 | $2 |
| Output $ / M tokens | $4.40 | $6 |
| Results tracked | 41 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.6 leads
GLM-5.1: 48.7 (#55), Grok 4.6: 58.5 (#16)
| Benchmark | GLM-5.1 | Grok 4.6 |
|---|---|---|
| LMArena WebDev | 1508 | 1617 |
| SciCode | 43.8% | 56.5% |
| WeirdML | 57.1% | 67.3% |
| LMArena Coding | 1485 | 1465 |
| ALE-Bench | 887.1 | 1,508 |
| SWE-bench Verified | 74.2% | — |
| DeepSWE | — | 67.5% |
| FrontierCode | — | 48% |
| CursorBench | — | 41.4% |
| FrontierSWE | — | 25.3% |
Agentic & Tool Use Grok 4.6 leads
GLM-5.1: 24.9 (#113), Grok 4.6: 39.4 (#27)
| Benchmark | GLM-5.1 | Grok 4.6 |
|---|---|---|
| APEX-Agents | 40.9% | 65.3% |
| Vending-Bench 2 | 5,634 | 9,047 |
| ExploitBench | 18.1% | — |
| GBAEval | 0% | — |
| GDP.pdf | — | 17.2% |
Reasoning Grok 4.6 leads
GLM-5.1: 39.1 (#60), Grok 4.6: 61.4 (#20)
| Benchmark | GLM-5.1 | Grok 4.6 |
|---|---|---|
| SimpleBench | 55.1% | 75.9% |
| NYT Connections (extended) | 77.7% | 80% |
| CritPt | 4.6% | 19.7% |
| Chess Puzzles | 19% | 40% |
| LMArena Hard Prompts | 1472 | 1447 |
| Epoch Capabilities Index | 149.84 | 156.44 |
| ARC-AGI-2 | — | 67.1% |
| ARC-AGI-1 | — | 87.5% |
| Thematic Generalization | 69.8% | — |
| EBR-Bench | — | 30.5% |
| Mystery Game Puzzles | — | 34% |
| DTBench | — | 97.3% |
| LMCA | — | 48.5% |
Math Grok 4.6 leads
GLM-5.1: 49.7 (#60), Grok 4.6: 67.0 (#24)
| Benchmark | GLM-5.1 | Grok 4.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 36.8% | 66% |
| OTIS Mock AIME 2024-2025 | 93.3% | 99.2% |
| ProofBench | 22.2% | 51% |
| LMArena Math | 1473 | 1423 |
| FrontierMath Tier 4 | — | 31.7% |
| MathArena Final-Answer Competitions | 67.1% | — |
| FrontierMath (Feb 2025 set) | 33.4% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge Grok 4.6 leads
GLM-5.1: 54.9 (#50), Grok 4.6: 63.3 (#20)
| Benchmark | GLM-5.1 | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 89.9% | 94% |
| SimpleQA Verified | 34% | 49.3% |
| LMArena Expert | 1476 | 1467 |
Multimodal Not comparable
GLM-5.1: —, Grok 4.6: 43.6 (#23)
| Benchmark | GLM-5.1 | Grok 4.6 |
|---|---|---|
| LMArena Vision | — | 1263 |
| Blueprint-Bench 2 | — | 33.2% |
| Furniture Assembly | — | 40% |
| LMArena Document | — | 1452 |
Multilingual GLM-5.1 leads
GLM-5.1: 55.0 (#36), Grok 4.6: 53.0 (#74)
| Benchmark | GLM-5.1 | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1447 | 1420 |
| LMArena Chinese | 1515 | 1480 |
| LMArena French | 1474 | 1461 |
| LMArena German | 1465 | 1431 |
| LMArena Japanese | 1434 | 1376 |
| LMArena Korean | 1418 | 1397 |
| LMArena Russian | 1454 | 1422 |
| LMArena Spanish | 1469 | 1404 |
Instruction Following Too close to call
GLM-5.1: 76.3 (#42), Grok 4.6: 75.4 (#63)
| Benchmark | GLM-5.1 | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1451 | 1431 |
Long Context Too close to call
GLM-5.1: 44.9 (#53), Grok 4.6: 44.5 (#66)
| Benchmark | GLM-5.1 | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1466 | 1454 |
Writing & Preference GLM-5.1 leads
GLM-5.1: 66.9 (#31), Grok 4.6: 62.3 (#80)
| Benchmark | GLM-5.1 | Grok 4.6 |
|---|---|---|
| LMArena Text | 1461 | 1428 |
| LMArena Creative Writing | 1453 | 1428 |
| LMArena Multi-Turn | 1472 | 1425 |
| EQ-Bench Creative Writing | 1592 | — |
Frequently asked questions
Is GLM-5.1 better than Grok 4.6?
Grok 4.6 is the stronger model overall, scoring 56.9 to 47.8 on the Noometry Index.
Which is cheaper, GLM-5.1 or Grok 4.6?
GLM-5.1 is cheaper. It lists at $1.40 per million input tokens and $4.40 per million output tokens; Grok 4.6 lists at $2 and $6.
Is GLM-5.1 or Grok 4.6 better for coding?
Grok 4.6 scores higher on coding benchmarks: 58.5 versus 48.7 in the Noometry coding category.
Which has the bigger context window?
Grok 4.6 does, with 500K tokens against 200K.
How many benchmarks do GLM-5.1 and Grok 4.6 share?
33 benchmarks have published results for both models. GLM-5.1 has 41 scored results on Noometry and Grok 4.6 has 49.