Model comparison
GLM-5.2 vs Qwen3.8 27B
GLM-5.2 is the stronger model overall, scoring 50.9 to 45.3 on the Noometry Index. Qwen3.8 27B costs 1.9× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
Alibaba (Qwen)Summary
- They share 27 benchmarks with published results for both. GLM-5.2 scores higher in 8 categories and Qwen3.8 27B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.2 leads 55.4 to 35.7.
- The biggest single-benchmark swing is NYT Connections (extended): 74.3% for GLM-5.2 and 54.5% for Qwen3.8 27B.
- Qwen3.8 27B is cheaper at $0.99 / $1.49 per million input/output tokens, against $1.40 / $4.40 for GLM-5.2.
- GLM-5.2 accepts more context: 1M tokens versus 262K.
Side by side
| GLM-5.2 | Qwen3.8 27B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 50.9 | 45.3 |
| Released | 2026-06-13 | 2026-08-14 |
| Weights | Open | Open |
| Context window | 1M | 262K |
| Max output | 131K | 33K |
| Input $ / M tokens | $1.40 | $0.99 |
| Output $ / M tokens | $4.40 | $1.49 |
| Results tracked | 50 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GLM-5.2: 50.9 (#37), Qwen3.8 27B: 50.3 (#42)
| Benchmark | GLM-5.2 | Qwen3.8 27B |
|---|---|---|
| LMArena WebDev | 1604 | 1592 |
| SciCode | 50.5% | 46.6% |
| LMArena Coding | 1484 | 1487 |
| SWE-bench Verified | 78.7% | — |
| DeepSWE | 43.8% | — |
| FrontierCode | 24.5% | — |
| WeirdML | 70.1% | — |
| ALE-Bench | 1,047 | — |
Agentic & Tool Use Not comparable
GLM-5.2: 31.6 (#63), Qwen3.8 27B: —
| Benchmark | GLM-5.2 | Qwen3.8 27B |
|---|---|---|
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 31.7% | — |
| GBAEval | 0% | — |
| Vending-Bench 2 | 8,314 | — |
Reasoning GLM-5.2 leads
GLM-5.2: 42.1 (#46), Qwen3.8 27B: 31.5 (#71)
| Benchmark | GLM-5.2 | Qwen3.8 27B |
|---|---|---|
| NYT Connections (extended) | 74.3% | 54.5% |
| CritPt | 20.9% | 5.4% |
| LMArena Hard Prompts | 1480 | 1459 |
| DTBench | 93.6% | 88% |
| LMCA | 45.8% | 41.4% |
| Surface Evolver Bench | 55.6% | 45% |
| Epoch Capabilities Index | 151.76 | 149.38 |
| ARC-AGI-2 | 22.8% | — |
| SimpleBench | 58.8% | — |
| Kagi LLM Benchmark | 62.6% | — |
| ARC-AGI-1 | 77% | — |
| Chess Puzzles | 21% | — |
| EBR-Bench | 9.5% | — |
| Mystery Game Puzzles | 19% | — |
Math GLM-5.2 leads
GLM-5.2: 55.4 (#35), Qwen3.8 27B: 35.7 (#139)
| Benchmark | GLM-5.2 | Qwen3.8 27B |
|---|---|---|
| ProofBench | 35% | 16% |
| LMArena Math | 1482 | 1453 |
| FrontierMath (Tiers 1-3) | 59.2% | — |
| FrontierMath Tier 4 | 29.3% | — |
| MathArena Final-Answer Competitions | 67.6% | — |
| OTIS Mock AIME 2024-2025 | 86.4% | — |
Knowledge GLM-5.2 leads
GLM-5.2: 57.1 (#36), Qwen3.8 27B: 40.9 (#105)
| Benchmark | GLM-5.2 | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | 1484 | 1482 |
| GPQA Diamond | 91.9% | — |
| SimpleQA Verified | 34.2% | — |
Multimodal Not comparable
GLM-5.2: —, Qwen3.8 27B: 40.5 (#36)
| Benchmark | GLM-5.2 | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | — | 1270 |
Multilingual GLM-5.2 leads
GLM-5.2: 56.0 (#25), Qwen3.8 27B: 54.0 (#54)
| Benchmark | GLM-5.2 | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1459 | 1431 |
| LMArena Chinese | 1518 | 1504 |
| LMArena French | 1476 | 1468 |
| LMArena German | 1470 | 1435 |
| LMArena Japanese | 1451 | 1392 |
| LMArena Korean | 1445 | 1389 |
| LMArena Russian | 1467 | 1421 |
| LMArena Spanish | 1477 | 1444 |
Instruction Following GLM-5.2 leads
GLM-5.2: 77.1 (#32), Qwen3.8 27B: 75.8 (#54)
| Benchmark | GLM-5.2 | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1465 | 1436 |
Long Context GLM-5.2 leads
GLM-5.2: 45.3 (#43), Qwen3.8 27B: 44.3 (#66)
| Benchmark | GLM-5.2 | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1479 | 1448 |
Writing & Preference GLM-5.2 leads
GLM-5.2: 70.7 (#20), Qwen3.8 27B: 65.9 (#41)
| Benchmark | GLM-5.2 | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1470 | 1441 |
| LMArena Creative Writing | 1465 | 1382 |
| EQ-Bench Creative Writing | 1757 | 1671 |
| LMArena Multi-Turn | 1468 | 1443 |
| EQ-Bench 4 | 1222 | — |
Frequently asked questions
Is GLM-5.2 better than Qwen3.8 27B?
GLM-5.2 is the stronger model overall, scoring 50.9 to 45.3 on the Noometry Index. Qwen3.8 27B costs 1.9× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Which is cheaper, GLM-5.2 or Qwen3.8 27B?
Qwen3.8 27B is cheaper. It lists at $0.99 per million input tokens and $1.49 per million output tokens; GLM-5.2 lists at $1.40 and $4.40.
Is GLM-5.2 or Qwen3.8 27B better for coding?
They score almost the same on coding (50.9 vs 50.3); test both on your own repository before choosing.
Which has the bigger context window?
GLM-5.2 does, with 1M tokens against 262K.
How many benchmarks do GLM-5.2 and Qwen3.8 27B share?
27 benchmarks have published results for both models. GLM-5.2 has 50 scored results on Noometry and Qwen3.8 27B has 28.