Model comparison
GLM-5 vs Qwen3.8 27B
GLM-5 and Qwen3.8 27B score almost the same on the Noometry Index (45.9 vs 45.3), so choose on price, context window or the category you care about most.
Last verified . 21 shared benchmarks.
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. GLM-5 scores higher in 4 categories and Qwen3.8 27B in 4 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-5 leads 52.2 to 40.9.
- The biggest single-benchmark swing is NYT Connections (extended): 74.8% for GLM-5 and 54.5% for Qwen3.8 27B.
- Qwen3.8 27B is cheaper at $0.99 / $1.49 per million input/output tokens, against $1 / $3.20 for GLM-5.
- Qwen3.8 27B accepts more context: 262K tokens versus 205K.
Side by side
| GLM-5 | Qwen3.8 27B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 45.9 | 45.3 |
| Released | 2026-02-11 | 2026-08-14 |
| Weights | Open | Open |
| Context window | 205K | 262K |
| Max output | 131K | 33K |
| Input $ / M tokens | $1 | $0.99 |
| Output $ / M tokens | $3.20 | $1.49 |
| Results tracked | 45 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
GLM-5: 49.0 (#47), Qwen3.8 27B: 50.3 (#42)
| Benchmark | GLM-5 | Qwen3.8 27B |
|---|---|---|
| LMArena WebDev | 1434 | 1592 |
| LMArena Coding | 1461 | 1487 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| SWE-bench Multilingual | 69.7% | — |
| SciCode | — | 46.6% |
| WeirdML | 48.2% | — |
| ALE-Bench | 765.62 | — |
Agentic & Tool Use Not comparable
GLM-5: 30.9 (#65), Qwen3.8 27B: —
| Benchmark | GLM-5 | Qwen3.8 27B |
|---|---|---|
| Terminal-Bench | 52.4% | — |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| Vending-Bench 2 | 4,432 | — |
Reasoning Qwen3.8 27B leads
GLM-5: 27.6 (#100), Qwen3.8 27B: 31.5 (#71)
| Benchmark | GLM-5 | Qwen3.8 27B |
|---|---|---|
| NYT Connections (extended) | 74.8% | 54.5% |
| LMArena Hard Prompts | 1451 | 1459 |
| Epoch Capabilities Index | 145.84 | 149.38 |
| ARC-AGI-2 | 4.9% | — |
| SimpleBench | 53.2% | — |
| Kagi LLM Benchmark | 75% | — |
| ARC-AGI-1 | 44.7% | — |
| CritPt | — | 5.4% |
| Chess Puzzles | 10% | — |
| DTBench | — | 88% |
| LMCA | — | 41.4% |
| Surface Evolver Bench | — | 45% |
| ForecastBench | 61 | — |
Math GLM-5 leads
GLM-5: 44.8 (#66), Qwen3.8 27B: 35.7 (#139)
| Benchmark | GLM-5 | Qwen3.8 27B |
|---|---|---|
| LMArena Math | 1440 | 1453 |
| MathArena Final-Answer Competitions | 65.7% | — |
| OTIS Mock AIME 2024-2025 | 80% | — |
| ProofBench | — | 16% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-5 leads
GLM-5: 52.2 (#59), Qwen3.8 27B: 40.9 (#105)
| Benchmark | GLM-5 | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | 1454 | 1482 |
| GPQA Diamond | 87.8% | — |
| Vectara Hallucination Rate | 10.1% | — |
Multimodal Not comparable
GLM-5: —, Qwen3.8 27B: 40.5 (#36)
| Benchmark | GLM-5 | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | — | 1270 |
Multilingual Too close to call
GLM-5: 53.9 (#57), Qwen3.8 27B: 54.0 (#54)
| Benchmark | GLM-5 | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1430 | 1431 |
| LMArena Chinese | 1511 | 1504 |
| LMArena French | 1455 | 1468 |
| LMArena German | 1445 | 1435 |
| LMArena Japanese | 1416 | 1392 |
| LMArena Korean | 1421 | 1389 |
| LMArena Russian | 1436 | 1421 |
| LMArena Spanish | 1452 | 1444 |
Instruction Following Too close to call
GLM-5: 75.4 (#64), Qwen3.8 27B: 75.8 (#54)
| Benchmark | GLM-5 | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1427 | 1436 |
Long Context Too close to call
GLM-5: 44.6 (#58), Qwen3.8 27B: 44.3 (#66)
| Benchmark | GLM-5 | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1446 | 1448 |
| CL-bench | 18.7% | — |
Writing & Preference Too close to call
GLM-5: 66.2 (#38), Qwen3.8 27B: 65.9 (#41)
| Benchmark | GLM-5 | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1446 | 1441 |
| LMArena Creative Writing | 1438 | 1382 |
| EQ-Bench Creative Writing | 1601 | 1671 |
| LMArena Multi-Turn | 1456 | 1443 |
Frequently asked questions
Is GLM-5 better than Qwen3.8 27B?
GLM-5 and Qwen3.8 27B score almost the same on the Noometry Index (45.9 vs 45.3), so choose on price, context window or the category you care about most.
Which is cheaper, GLM-5 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 lists at $1 and $3.20.
Is GLM-5 or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.3 versus 49.0 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 27B does, with 262K tokens against 205K.
How many benchmarks do GLM-5 and Qwen3.8 27B share?
21 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and Qwen3.8 27B has 28.