Model comparison
GLM-4.7 vs Grok 4.6
Grok 4.6 is the stronger model overall, scoring 56.9 to 42.0 on the Noometry Index. GLM-4.7 costs 3.0× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. GLM-4.7 scores higher in 0 categories and Grok 4.6 in 9 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 24.3.
- The biggest single-benchmark swing is ProofBench: 6% for GLM-4.7 and 51% for Grok 4.6.
- GLM-4.7 is cheaper at $0.60 / $2.20 per million input/output tokens, against $2 / $6 for Grok 4.6.
- Grok 4.6 accepts more context: 500K tokens versus 205K.
- GLM-4.7 has downloadable open weights; the other is API-only.
Side by side
| GLM-4.7 | Grok 4.6 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 42.0 | 56.9 |
| Released | 2025-12-22 | 2026-08-12 |
| Weights | Open | Proprietary |
| Context window | 205K | 500K |
| Max output | 131K | 500K |
| Input $ / M tokens | $0.60 | $2 |
| Output $ / M tokens | $2.20 | $6 |
| Results tracked | 36 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.6 leads
GLM-4.7: 44.0 (#79), Grok 4.6: 58.5 (#16)
| Benchmark | GLM-4.7 | Grok 4.6 |
|---|---|---|
| LMArena WebDev | 1435 | 1617 |
| SciCode | 45.1% | 56.5% |
| LMArena Coding | 1454 | 1465 |
| ALE-Bench | 399.48 | 1,508 |
| DeepSWE | — | 67.5% |
| FrontierCode | — | 48% |
| CursorBench | — | 41.4% |
| FrontierSWE | — | 25.3% |
| WeirdML | — | 67.3% |
Agentic & Tool Use Grok 4.6 leads
GLM-4.7: 26.5 (#103), Grok 4.6: 39.4 (#27)
| Benchmark | GLM-4.7 | Grok 4.6 |
|---|---|---|
| Vending-Bench 2 | 2,377 | 9,047 |
| Terminal-Bench | 33.4% | — |
| APEX-Agents | — | 65.3% |
| GDP.pdf | — | 17.2% |
Reasoning Grok 4.6 leads
GLM-4.7: 24.3 (#164), Grok 4.6: 61.4 (#20)
| Benchmark | GLM-4.7 | Grok 4.6 |
|---|---|---|
| SimpleBench | 47.7% | 75.9% |
| CritPt | 1.7% | 19.7% |
| Chess Puzzles | 6% | 40% |
| LMArena Hard Prompts | 1443 | 1447 |
| Epoch Capabilities Index | 143.51 | 156.44 |
| ARC-AGI-2 | — | 67.1% |
| NYT Connections (extended) | — | 80% |
| ARC-AGI-1 | — | 87.5% |
| EBR-Bench | — | 30.5% |
| Mystery Game Puzzles | — | 34% |
| DTBench | — | 97.3% |
| LMCA | — | 48.5% |
Math Grok 4.6 leads
GLM-4.7: 38.6 (#135), Grok 4.6: 67.0 (#24)
| Benchmark | GLM-4.7 | Grok 4.6 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 83.3% | 99.2% |
| ProofBench | 6% | 51% |
| LMArena Math | 1423 | 1423 |
| FrontierMath (Tiers 1-3) | — | 66% |
| FrontierMath Tier 4 | — | 31.7% |
| FrontierMath (Feb 2025 set) | 2.4% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Grok 4.6 leads
GLM-4.7: 47.0 (#80), Grok 4.6: 63.3 (#20)
| Benchmark | GLM-4.7 | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 83.3% | 94% |
| SimpleQA Verified | 32.2% | 49.3% |
| LMArena Expert | 1424 | 1467 |
| Vectara Hallucination Rate | 11.7% | — |
Multimodal Not comparable
GLM-4.7: —, Grok 4.6: 43.6 (#23)
| Benchmark | GLM-4.7 | Grok 4.6 |
|---|---|---|
| LMArena Vision | — | 1263 |
| Blueprint-Bench 2 | — | 33.2% |
| Furniture Assembly | — | 40% |
| LMArena Document | — | 1452 |
Multilingual Too close to call
GLM-4.7: 52.8 (#79), Grok 4.6: 53.0 (#74)
| Benchmark | GLM-4.7 | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1417 | 1420 |
| LMArena Chinese | 1495 | 1480 |
| LMArena French | 1432 | 1461 |
| LMArena German | 1424 | 1431 |
| LMArena Japanese | 1439 | 1376 |
| LMArena Korean | 1399 | 1397 |
| LMArena Russian | 1423 | 1422 |
| LMArena Spanish | 1434 | 1404 |
Instruction Following Too close to call
GLM-4.7: 74.4 (#95), Grok 4.6: 75.4 (#63)
| Benchmark | GLM-4.7 | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1411 | 1431 |
Long Context Grok 4.6 leads
GLM-4.7: 42.8 (#116), Grok 4.6: 44.5 (#66)
| Benchmark | GLM-4.7 | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1432 | 1454 |
| CL-bench | 15.9% | — |
| CL-bench Life | 10.9% | — |
Writing & Preference Grok 4.6 leads
GLM-4.7: 60.9 (#93), Grok 4.6: 62.3 (#80)
| Benchmark | GLM-4.7 | Grok 4.6 |
|---|---|---|
| LMArena Text | 1435 | 1428 |
| LMArena Creative Writing | 1401 | 1428 |
| LMArena Multi-Turn | 1446 | 1425 |
| EQ-Bench Creative Writing | 1413 | — |
Frequently asked questions
Is GLM-4.7 better than Grok 4.6?
Grok 4.6 is the stronger model overall, scoring 56.9 to 42.0 on the Noometry Index. GLM-4.7 costs 3.0× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Which is cheaper, GLM-4.7 or Grok 4.6?
GLM-4.7 is cheaper. It lists at $0.60 per million input tokens and $2.20 per million output tokens; Grok 4.6 lists at $2 and $6.
Is GLM-4.7 or Grok 4.6 better for coding?
Grok 4.6 scores higher on coding benchmarks: 58.5 versus 44.0 in the Noometry coding category.
Which has the bigger context window?
Grok 4.6 does, with 500K tokens against 205K.
How many benchmarks do GLM-4.7 and Grok 4.6 share?
29 benchmarks have published results for both models. GLM-4.7 has 36 scored results on Noometry and Grok 4.6 has 49.