Model comparison
GLM-4.6 vs Grok 4.3
Grok 4.3 is the stronger model overall, scoring 43.6 to 40.9 on the Noometry Index. GLM-4.6 costs 1.6× less per token, which makes it the better buy when Grok 4.3's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
Summary
- They share 21 benchmarks with published results for both. GLM-4.6 scores higher in 5 categories and Grok 4.3 in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok 4.3 leads 52.5 to 39.6.
- The biggest single-benchmark swing is SciCode: 38.4% for GLM-4.6 and 47.3% for Grok 4.3.
- GLM-4.6 is cheaper at $0.60 / $2.20 per million input/output tokens, against $1.25 / $2.50 for Grok 4.3.
- Grok 4.3 accepts more context: 1M tokens versus 205K.
- GLM-4.6 has downloadable open weights; the other is API-only.
Side by side
| GLM-4.6 | Grok 4.3 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 40.9 | 43.6 |
| Released | 2025-09-30 | 2026-04-17 |
| Weights | Open | Proprietary |
| Context window | 205K | 1M |
| Max output | 131K | 30K |
| Input $ / M tokens | $0.60 | $1.25 |
| Output $ / M tokens | $2.20 | $2.50 |
| Results tracked | 29 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.3 leads
GLM-4.6: 39.9 (#137), Grok 4.3: 41.3 (#117)
| Benchmark | GLM-4.6 | Grok 4.3 |
|---|---|---|
| LMArena WebDev | 1339 | 1357 |
| SciCode | 38.4% | 47.3% |
| LMArena Coding | 1449 | 1415 |
| ALE-Bench | 340.82 | 944.17 |
| SWE-bench Verified (bash only) | 55.4% | — |
| WeirdML | — | 49.9% |
Agentic & Tool Use GLM-4.6 leads
GLM-4.6: 32.1 (#60), Grok 4.3: 27.6 (#93)
| Benchmark | GLM-4.6 | Grok 4.3 |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| Berkeley Function Calling Leaderboard | 72.4% | — |
| GDP.pdf | — | 8% |
| LMArena Search | — | 1165 |
| Vending-Bench 2 | — | 35.26 |
Reasoning Grok 4.3 leads
GLM-4.6: 23.1 (#168), Grok 4.3: 34.9 (#60)
| Benchmark | GLM-4.6 | Grok 4.3 |
|---|---|---|
| CritPt | 1.1% | 8% |
| LMArena Hard Prompts | 1440 | 1397 |
| Kagi LLM Benchmark | 47.4% | — |
| NYT Connections (extended) | — | 55.2% |
| Chess Puzzles | — | 25% |
| DTBench | — | 90.7% |
| LMCA | — | 38.3% |
| Epoch Capabilities Index | — | 149.13 |
| ForecastBench | — | 60.3 |
Math Grok 4.3 leads
GLM-4.6: 36.4 (#115), Grok 4.3: 45.8 (#60)
| Benchmark | GLM-4.6 | Grok 4.3 |
|---|---|---|
| LMArena Math | 1431 | 1388 |
| FrontierMath (Tiers 1-3) | — | 42.8% |
| FrontierMath Tier 4 | — | 14.6% |
| OTIS Mock AIME 2024-2025 | — | 93.3% |
| ProofBench | — | 11% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Grok 4.3 leads
GLM-4.6: 39.6 (#118), Grok 4.3: 52.5 (#57)
| Benchmark | GLM-4.6 | Grok 4.3 |
|---|---|---|
| LMArena Expert | 1432 | 1385 |
| GPQA Diamond | — | 88.8% |
| SimpleQA Verified | — | 33.2% |
| Vectara Hallucination Rate | 9.5% | — |
Multimodal Not comparable
GLM-4.6: —, Grok 4.3: 31.6 (#96)
| Benchmark | GLM-4.6 | Grok 4.3 |
|---|---|---|
| LMArena Vision | — | 1231 |
| Blueprint-Bench 2 | — | 0% |
Multilingual GLM-4.6 leads
GLM-4.6: 53.6 (#62), Grok 4.3: 50.7 (#114)
| Benchmark | GLM-4.6 | Grok 4.3 |
|---|---|---|
| LMArena Non-English | 1426 | 1385 |
| LMArena Chinese | 1500 | 1423 |
| LMArena French | 1459 | 1412 |
| LMArena German | 1447 | 1395 |
| LMArena Japanese | 1394 | 1379 |
| LMArena Korean | 1401 | 1357 |
| LMArena Russian | 1418 | 1399 |
| LMArena Spanish | 1436 | 1402 |
Instruction Following GLM-4.6 leads
GLM-4.6: 74.5 (#93), Grok 4.3: 72.3 (#135)
| Benchmark | GLM-4.6 | Grok 4.3 |
|---|---|---|
| LMArena Instruction Following | 1410 | 1366 |
Long Context Too close to call
GLM-4.6: 43.4 (#91), Grok 4.3: 42.5 (#118)
| Benchmark | GLM-4.6 | Grok 4.3 |
|---|---|---|
| LMArena Longer Query | 1423 | 1393 |
Writing & Preference GLM-4.6 leads
GLM-4.6: 61.3 (#86), Grok 4.3: 58.6 (#112)
| Benchmark | GLM-4.6 | Grok 4.3 |
|---|---|---|
| LMArena Text | 1440 | 1397 |
| LMArena Creative Writing | 1411 | 1380 |
| LMArena Multi-Turn | 1427 | 1406 |
| EQ-Bench Creative Writing | 1411 | — |
| EQ-Bench 4 | — | 1075 |
Frequently asked questions
Is GLM-4.6 better than Grok 4.3?
Grok 4.3 is the stronger model overall, scoring 43.6 to 40.9 on the Noometry Index. GLM-4.6 costs 1.6× less per token, which makes it the better buy when Grok 4.3's lead doesn't matter for your workload.
Which is cheaper, GLM-4.6 or Grok 4.3?
GLM-4.6 is cheaper. It lists at $0.60 per million input tokens and $2.20 per million output tokens; Grok 4.3 lists at $1.25 and $2.50.
Is GLM-4.6 or Grok 4.3 better for coding?
Grok 4.3 scores higher on coding benchmarks: 41.3 versus 39.9 in the Noometry coding category.
Which has the bigger context window?
Grok 4.3 does, with 1M tokens against 205K.
How many benchmarks do GLM-4.6 and Grok 4.3 share?
21 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Grok 4.3 has 40.