Model comparison
GLM-4.6 vs Qwen3-4B
GLM-4.6 is the stronger model overall, scoring 41.4 to 31.9 on the Noometry Index.
Last verified . 2 shared benchmarks.
Alibaba (Qwen)Summary
- They share 2 benchmarks with published results for both. GLM-4.6 scores higher in 4 categories and Qwen3-4B in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-4.6 leads 39.1 to 29.7.
- The biggest single-benchmark swing is Berkeley Function Calling Leaderboard: 72.4% for GLM-4.6 and 35.7% for Qwen3-4B.
Side by side
| GLM-4.6 | Qwen3-4B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 41.4 | 31.9 |
| Released | 2025-09-30 | 2025-04-29 |
| Weights | Open | Open |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.60 | — |
| Output $ / M tokens | $2.20 | — |
| Results tracked | 29 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
GLM-4.6: 40.1 (#148), Qwen3-4B: —
| Benchmark | GLM-4.6 | Qwen3-4B |
|---|---|---|
| SWE-bench Verified (bash only) | 55.4% | — |
| LMArena WebDev | 1340 | — |
| SciCode | 38.4% | — |
| LMArena Coding | 1449 | — |
| ALE-Bench | 340.82 | — |
Agentic & Tool Use GLM-4.6 leads
GLM-4.6: 32.3 (#66), Qwen3-4B: 27.6 (#100)
| Benchmark | GLM-4.6 | Qwen3-4B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 72.4% | 35.7% |
| Terminal-Bench | 24.5% | — |
Reasoning GLM-4.6 leads
GLM-4.6: 23.7 (#172), Qwen3-4B: 19.2 (#268)
| Benchmark | GLM-4.6 | Qwen3-4B |
|---|---|---|
| Kagi LLM Benchmark | 47.4% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | — | 4% |
| LMArena Hard Prompts | 1440 | — |
Math GLM-4.6 leads
GLM-4.6: 39.1 (#111), Qwen3-4B: 29.7 (#240)
| Benchmark | GLM-4.6 | Qwen3-4B |
|---|---|---|
| MathArena Final-Answer Competitions | — | 38.5% |
| OTIS Mock AIME 2024-2025 | — | 52.2% |
| LMArena Math | 1432 | — |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-4.6 leads
GLM-4.6: 40.2 (#124), Qwen3-4B: 33.0 (#208)
| Benchmark | GLM-4.6 | Qwen3-4B |
|---|---|---|
| Vectara Hallucination Rate | 9.5% | 5.7% |
| GPQA Diamond | — | 52.3% |
| LMArena Expert | 1431 | — |
Multilingual Not comparable
GLM-4.6: 53.5 (#66), Qwen3-4B: —
| Benchmark | GLM-4.6 | Qwen3-4B |
|---|---|---|
| LMArena Non-English | 1426 | — |
| LMArena Chinese | 1499 | — |
| LMArena French | 1459 | — |
| LMArena German | 1447 | — |
| LMArena Japanese | 1393 | — |
| LMArena Korean | 1400 | — |
| LMArena Russian | 1419 | — |
| LMArena Spanish | 1436 | — |
Instruction Following Not comparable
GLM-4.6: 74.3 (#98), Qwen3-4B: —
| Benchmark | GLM-4.6 | Qwen3-4B |
|---|---|---|
| LMArena Instruction Following | 1410 | — |
Long Context Not comparable
GLM-4.6: 43.4 (#94), Qwen3-4B: —
| Benchmark | GLM-4.6 | Qwen3-4B |
|---|---|---|
| LMArena Longer Query | 1422 | — |
Writing & Preference Not comparable
GLM-4.6: 61.1 (#90), Qwen3-4B: —
| Benchmark | GLM-4.6 | Qwen3-4B |
|---|---|---|
| LMArena Text | 1440 | — |
| LMArena Creative Writing | 1411 | — |
| EQ-Bench Creative Writing | 1411 | — |
| LMArena Multi-Turn | 1427 | — |
Frequently asked questions
Is GLM-4.6 better than Qwen3-4B?
GLM-4.6 is the stronger model overall, scoring 41.4 to 31.9 on the Noometry Index.
How many benchmarks do GLM-4.6 and Qwen3-4B share?
2 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and Qwen3-4B has 6.