Model comparison
GLM-5 vs Qwen3-30B-A3B
GLM-5 is the stronger model overall, scoring 45.9 to 38.5 on the Noometry Index. Qwen3-30B-A3B costs 7.2× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. GLM-5 scores higher in 9 categories and Qwen3-30B-A3B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in long context, where GLM-5 leads 44.6 to 31.0.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 75% for GLM-5 and 54.9% for Qwen3-30B-A3B.
- Qwen3-30B-A3B is cheaper at $0.12 / $0.50 per million input/output tokens, against $1 / $3.20 for GLM-5.
- GLM-5 accepts more context: 205K tokens versus 41K.
Side by side
| GLM-5 | Qwen3-30B-A3B | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 45.9 | 38.5 |
| Released | 2026-02-11 | 2025-04-28 |
| Weights | Open | Open |
| Context window | 205K | 41K |
| Max output | 131K | 16K |
| Input $ / M tokens | $1 | $0.12 |
| Output $ / M tokens | $3.20 | $0.50 |
| Results tracked | 45 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
GLM-5: 49.0 (#47), Qwen3-30B-A3B: 37.0 (#188)
| Benchmark | GLM-5 | Qwen3-30B-A3B |
|---|---|---|
| WeirdML | 48.2% | 29.8% |
| LMArena Coding | 1461 | 1416 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| SciCode | — | 33.3% |
| ALE-Bench | 765.62 | — |
Agentic & Tool Use GLM-5 leads
GLM-5: 30.9 (#65), Qwen3-30B-A3B: 29.5 (#75)
| Benchmark | GLM-5 | Qwen3-30B-A3B |
|---|---|---|
| Terminal-Bench | 52.4% | — |
| Berkeley Function Calling Leaderboard | — | 41.4% |
| 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 GLM-5 leads
GLM-5: 27.6 (#100), Qwen3-30B-A3B: 21.6 (#193)
| Benchmark | GLM-5 | Qwen3-30B-A3B |
|---|---|---|
| Kagi LLM Benchmark | 75% | 54.9% |
| Chess Puzzles | 10% | 8% |
| LMArena Hard Prompts | 1451 | 1398 |
| Epoch Capabilities Index | 145.84 | 139.63 |
| ARC-AGI-2 | 4.9% | — |
| SimpleBench | 53.2% | — |
| NYT Connections (extended) | 74.8% | — |
| ARC-AGI-1 | 44.7% | — |
| CritPt | — | 0.3% |
| DTBench | — | 69.3% |
| LMCA | — | 22.4% |
| ForecastBench | 61 | — |
Math GLM-5 leads
GLM-5: 44.8 (#66), Qwen3-30B-A3B: 36.0 (#133)
| Benchmark | GLM-5 | Qwen3-30B-A3B |
|---|---|---|
| MathArena Final-Answer Competitions | 65.7% | 47.8% |
| OTIS Mock AIME 2024-2025 | 80% | 70.3% |
| LMArena Math | 1440 | 1394 |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-5 leads
GLM-5: 52.2 (#59), Qwen3-30B-A3B: 41.7 (#96)
| Benchmark | GLM-5 | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 87.8% | 70.1% |
| LMArena Expert | 1454 | 1395 |
| Confabulations | — | 12.3% |
| Vectara Hallucination Rate | 10.1% | — |
Multilingual GLM-5 leads
GLM-5: 53.9 (#57), Qwen3-30B-A3B: 49.8 (#126)
| Benchmark | GLM-5 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1430 | 1372 |
| LMArena Chinese | 1511 | 1434 |
| LMArena French | 1455 | 1419 |
| LMArena German | 1445 | 1379 |
| LMArena Japanese | 1416 | 1338 |
| LMArena Korean | 1421 | 1333 |
| LMArena Russian | 1436 | 1369 |
| LMArena Spanish | 1452 | 1403 |
Instruction Following GLM-5 leads
GLM-5: 75.4 (#64), Qwen3-30B-A3B: 72.2 (#137)
| Benchmark | GLM-5 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1427 | 1364 |
Long Context GLM-5 leads
GLM-5: 44.6 (#58), Qwen3-30B-A3B: 31.0 (#278)
| Benchmark | GLM-5 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Longer Query | 1446 | 1379 |
| Fiction.LiveBench | — | 40.6% |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.2 (#38), Qwen3-30B-A3B: 55.9 (#138)
| Benchmark | GLM-5 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1446 | 1384 |
| LMArena Creative Writing | 1438 | 1318 |
| LMArena Multi-Turn | 1456 | 1378 |
| Short-Story Creative Writing | — | 75.3% |
| EQ-Bench Creative Writing | 1601 | — |
Frequently asked questions
Is GLM-5 better than Qwen3-30B-A3B?
GLM-5 is the stronger model overall, scoring 45.9 to 38.5 on the Noometry Index. Qwen3-30B-A3B costs 7.2× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Which is cheaper, GLM-5 or Qwen3-30B-A3B?
Qwen3-30B-A3B is cheaper. It lists at $0.12 per million input tokens and $0.50 per million output tokens; GLM-5 lists at $1 and $3.20.
Is GLM-5 or Qwen3-30B-A3B better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 37.0 in the Noometry coding category.
Which has the bigger context window?
GLM-5 does, with 205K tokens against 41K.
How many benchmarks do GLM-5 and Qwen3-30B-A3B share?
24 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and Qwen3-30B-A3B has 32.