Model comparison
GLM-5 vs Qwen3 Max
GLM-5 is the stronger model overall, scoring 46.1 to 43.7 on the Noometry Index.
Last verified . 26 shared benchmarks.
Alibaba (Qwen)Summary
- They share 26 benchmarks with published results for both. GLM-5 scores higher in 8 categories and Qwen3 Max in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5 leads 46.4 to 38.7.
- The biggest single-benchmark swing is NYT Connections (extended): 74.8% for GLM-5 and 30.1% for Qwen3 Max.
- GLM-5 is cheaper at $1 / $3.20 per million input/output tokens, against $1.20 / $6 for Qwen3 Max.
- Qwen3 Max accepts more context: 262K tokens versus 205K.
- GLM-5 has downloadable open weights; the other is API-only.
Side by side
| GLM-5 | Qwen3 Max | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 46.1 | 43.7 |
| Released | 2026-02-11 | 2025-09-23 |
| Weights | Open | Proprietary |
| Context window | 205K | 262K |
| Max output | 131K | 66K |
| Input $ / M tokens | $1 | $1.20 |
| Output $ / M tokens | $3.20 | $6 |
| Results tracked | 45 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
GLM-5: 49.0 (#52), Qwen3 Max: 43.0 (#93)
| Benchmark | GLM-5 | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1461 | 1456 |
| ALE-Bench | 765.62 | 370.45 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| WeirdML | 48.2% | — |
Agentic & Tool Use Not comparable
GLM-5: 31.1 (#71), Qwen3 Max: —
| Benchmark | GLM-5 | Qwen3 Max |
|---|---|---|
| Vending-Bench 2 | 4,432 | 71.56 |
| Terminal-Bench | 52.4% | — |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
Reasoning GLM-5 leads
GLM-5: 27.6 (#116), Qwen3 Max: 22.6 (#190)
| Benchmark | GLM-5 | Qwen3 Max |
|---|---|---|
| Kagi LLM Benchmark | 75% | 72.5% |
| NYT Connections (extended) | 74.8% | 30.1% |
| Chess Puzzles | 10% | 4% |
| LMArena Hard Prompts | 1452 | 1448 |
| Epoch Capabilities Index | 145.83 | 142.38 |
| ARC-AGI-2 | 4.9% | — |
| SimpleBench | 53.2% | — |
| ARC-AGI-1 | 44.7% | — |
| Mystery Game Puzzles | — | 5% |
| DTBench | — | 82.1% |
| LMCA | — | 28.3% |
| ForecastBench | 61 | — |
Math GLM-5 leads
GLM-5: 46.4 (#71), Qwen3 Max: 38.7 (#131)
| Benchmark | GLM-5 | Qwen3 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 73.3% |
| LMArena Math | 1440 | 1446 |
| FrontierMath (Tiers 1-3) | — | 18.9% |
| MathArena Final-Answer Competitions | 65.7% | — |
| MATH Level 5 | — | 97.1% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-5 leads
GLM-5: 52.3 (#64), Qwen3 Max: 48.1 (#78)
| Benchmark | GLM-5 | Qwen3 Max |
|---|---|---|
| GPQA Diamond | 87.8% | 72.6% |
| LMArena Expert | 1454 | 1455 |
| SimpleQA Verified | — | 48.7% |
| Vectara Hallucination Rate | 10.1% | — |
Multilingual Too close to call
GLM-5: 53.7 (#58), Qwen3 Max: 53.7 (#62)
| Benchmark | GLM-5 | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1430 | 1429 |
| LMArena Chinese | 1511 | 1478 |
| LMArena French | 1455 | 1449 |
| LMArena German | 1445 | 1463 |
| LMArena Japanese | 1416 | 1397 |
| LMArena Korean | 1423 | 1399 |
| LMArena Russian | 1436 | 1428 |
| LMArena Spanish | 1454 | 1462 |
Instruction Following Too close to call
GLM-5: 75.2 (#67), Qwen3 Max: 74.8 (#87)
| Benchmark | GLM-5 | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1428 | 1419 |
Long Context GLM-5 leads
GLM-5: 44.7 (#60), Qwen3 Max: 41.6 (#134)
| Benchmark | GLM-5 | Qwen3 Max |
|---|---|---|
| CL-bench | 18.7% | 14.5% |
| LMArena Longer Query | 1446 | 1438 |
| Fiction.LiveBench | — | 66.7% |
Writing & Preference GLM-5 leads
GLM-5: 66.0 (#38), Qwen3 Max: 62.4 (#76)
| Benchmark | GLM-5 | Qwen3 Max |
|---|---|---|
| LMArena Text | 1446 | 1439 |
| LMArena Creative Writing | 1439 | 1402 |
| LMArena Multi-Turn | 1456 | 1446 |
| EQ-Bench Creative Writing | 1601 | — |
Frequently asked questions
Is GLM-5 better than Qwen3 Max?
GLM-5 is the stronger model overall, scoring 46.1 to 43.7 on the Noometry Index.
Which is cheaper, GLM-5 or Qwen3 Max?
GLM-5 is cheaper. It lists at $1 per million input tokens and $3.20 per million output tokens; Qwen3 Max lists at $1.20 and $6.
Is GLM-5 or Qwen3 Max better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 43.0 in the Noometry coding category.
Which has the bigger context window?
Qwen3 Max does, with 262K tokens against 205K.
How many benchmarks do GLM-5 and Qwen3 Max share?
26 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and Qwen3 Max has 33.