Model comparison
GLM-5.1 vs Qwen3.6 27B
GLM-5.1 is the stronger model overall, scoring 47.6 to 41.9 on the Noometry Index. Qwen3.6 27B costs 1.6× less per token, which makes it the better buy when GLM-5.1's lead doesn't matter for your workload.
Last verified . 7 shared benchmarks.
Alibaba (Qwen)Summary
- They share 7 benchmarks with published results for both. GLM-5.1 scores higher in 5 categories and Qwen3.6 27B in 0 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-5.1 leads 67.2 to 50.3.
- The biggest single-benchmark swing is SciCode: 43.8% for GLM-5.1 and 37.3% for Qwen3.6 27B.
- Qwen3.6 27B is cheaper at $0.60 / $3.60 per million input/output tokens, against $1.40 / $4.40 for GLM-5.1.
- Qwen3.6 27B accepts more context: 262K tokens versus 200K.
Side by side
| GLM-5.1 | Qwen3.6 27B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 47.6 | 41.9 |
| Released | 2026-04-07 | 2026-04-22 |
| Weights | Open | Open |
| Context window | 200K | 262K |
| Max output | 131K | 66K |
| Input $ / M tokens | $1.40 | $0.60 |
| Output $ / M tokens | $4.40 | $3.60 |
| Results tracked | 41 | 11 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.1 leads
GLM-5.1: 48.5 (#52), Qwen3.6 27B: 38.5 (#162)
| Benchmark | GLM-5.1 | Qwen3.6 27B |
|---|---|---|
| SciCode | 43.8% | 37.3% |
| SWE-bench Verified | 74.2% | — |
| LMArena WebDev | 1509 | — |
| WeirdML | 57.1% | — |
| LMArena Coding | 1486 | — |
| ALE-Bench | 887.1 | — |
Agentic & Tool Use Not comparable
GLM-5.1: 24.8 (#106), Qwen3.6 27B: —
| Benchmark | GLM-5.1 | Qwen3.6 27B |
|---|---|---|
| APEX-Agents | 40.9% | — |
| ExploitBench | 18.1% | — |
| GBAEval | 0% | — |
| Vending-Bench 2 | 5,634 | — |
Reasoning GLM-5.1 leads
GLM-5.1: 38.2 (#57), Qwen3.6 27B: 24.4 (#148)
| Benchmark | GLM-5.1 | Qwen3.6 27B |
|---|---|---|
| CritPt | 4.6% | 0.9% |
| Chess Puzzles | 19% | 22% |
| Epoch Capabilities Index | 149.86 | 146.47 |
| SimpleBench | 55.1% | — |
| NYT Connections (extended) | 77.7% | — |
| Thematic Generalization | 69.8% | — |
| LMArena Hard Prompts | 1472 | — |
| Mystery Game Puzzles | — | 7% |
| DTBench | — | 78.1% |
| LMCA | — | 34.5% |
Math GLM-5.1 leads
GLM-5.1: 49.0 (#48), Qwen3.6 27B: 47.5 (#56)
| Benchmark | GLM-5.1 | Qwen3.6 27B |
|---|---|---|
| FrontierMath (Tiers 1-3) | 36.8% | 35.1% |
| OTIS Mock AIME 2024-2025 | 93.3% | 91.1% |
| MathArena Final-Answer Competitions | 67.1% | — |
| ProofBench | 22.2% | — |
| LMArena Math | 1472 | — |
| FrontierMath (Feb 2025 set) | 33.4% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge GLM-5.1 leads
GLM-5.1: 55.0 (#45), Qwen3.6 27B: 52.4 (#58)
| Benchmark | GLM-5.1 | Qwen3.6 27B |
|---|---|---|
| GPQA Diamond | 89.9% | 85.9% |
| SimpleQA Verified | 34% | — |
| LMArena Expert | 1477 | — |
Multilingual Not comparable
GLM-5.1: 55.2 (#34), Qwen3.6 27B: —
| Benchmark | GLM-5.1 | Qwen3.6 27B |
|---|---|---|
| LMArena Non-English | 1448 | — |
| LMArena Chinese | 1516 | — |
| LMArena French | 1478 | — |
| LMArena German | 1466 | — |
| LMArena Japanese | 1436 | — |
| LMArena Korean | 1423 | — |
| LMArena Russian | 1453 | — |
| LMArena Spanish | 1467 | — |
Instruction Following Not comparable
GLM-5.1: 76.5 (#41), Qwen3.6 27B: —
| Benchmark | GLM-5.1 | Qwen3.6 27B |
|---|---|---|
| LMArena Instruction Following | 1451 | — |
Long Context Not comparable
GLM-5.1: 44.9 (#51), Qwen3.6 27B: —
| Benchmark | GLM-5.1 | Qwen3.6 27B |
|---|---|---|
| LMArena Longer Query | 1467 | — |
Writing & Preference GLM-5.1 leads
GLM-5.1: 67.2 (#29), Qwen3.6 27B: 50.3 (#180)
| Benchmark | GLM-5.1 | Qwen3.6 27B |
|---|---|---|
| LMArena Text | 1462 | — |
| LMArena Creative Writing | 1454 | — |
| EQ-Bench Creative Writing | 1592 | — |
| EQ-Bench 4 | — | 1026 |
| LMArena Multi-Turn | 1473 | — |
Frequently asked questions
Is GLM-5.1 better than Qwen3.6 27B?
GLM-5.1 is the stronger model overall, scoring 47.6 to 41.9 on the Noometry Index. Qwen3.6 27B costs 1.6× less per token, which makes it the better buy when GLM-5.1's lead doesn't matter for your workload.
Which is cheaper, GLM-5.1 or Qwen3.6 27B?
Qwen3.6 27B is cheaper. It lists at $0.60 per million input tokens and $3.60 per million output tokens; GLM-5.1 lists at $1.40 and $4.40.
Is GLM-5.1 or Qwen3.6 27B better for coding?
GLM-5.1 scores higher on coding benchmarks: 48.5 versus 38.5 in the Noometry coding category.
Which has the bigger context window?
Qwen3.6 27B does, with 262K tokens against 200K.
How many benchmarks do GLM-5.1 and Qwen3.6 27B share?
7 benchmarks have published results for both models. GLM-5.1 has 41 scored results on Noometry and Qwen3.6 27B has 11.