Model comparison
GLM-5 vs Qwen2.5 72B Instruct
GLM-5 is the stronger model overall, scoring 46.1 to 31.9 on the Noometry Index.
Last verified . 22 shared benchmarks.
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. GLM-5 scores higher in 9 categories and Qwen2.5 72B Instruct in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5 leads 46.4 to 19.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 80% for GLM-5 and 8.1% for Qwen2.5 72B Instruct.
- GLM-5 is cheaper at $1 / $3.20 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- GLM-5 accepts more context: 205K tokens versus 131K.
Side by side
| GLM-5 | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Alibaba (Qwen) |
| Noometry Index | 46.1 | 31.9 |
| Released | 2026-02-11 | 2024-09 |
| Weights | Open | Open |
| Context window | 205K | 131K |
| Max output | 131K | 8K |
| Input $ / M tokens | $1 | $1.40 |
| Output $ / M tokens | $3.20 | $5.60 |
| Results tracked | 45 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
GLM-5: 49.0 (#52), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | GLM-5 | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 48.2% | 16% |
| LMArena Coding | 1461 | 1292 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| LMArena WebDev | 1434 | — |
| SWE-bench Multilingual | 69.7% | — |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
| ALE-Bench | 765.62 | — |
Agentic & Tool Use GLM-5 leads
GLM-5: 31.1 (#71), Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | GLM-5 | Qwen2.5 72B Instruct |
|---|---|---|
| Terminal-Bench | 52.4% | — |
| TheAgentCompany | — | 5.7% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
| Vending-Bench 2 | 4,432 | — |
Reasoning GLM-5 leads
GLM-5: 27.6 (#116), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | GLM-5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1452 | 1271 |
| Epoch Capabilities Index | 145.83 | 129 |
| ForecastBench | 61 | 57.5 |
| ARC-AGI-2 | 4.9% | — |
| SimpleBench | 53.2% | — |
| Kagi LLM Benchmark | 75% | — |
| NYT Connections (extended) | 74.8% | — |
| ARC-AGI-1 | 44.7% | — |
| Chess Puzzles | 10% | — |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math GLM-5 leads
GLM-5: 46.4 (#71), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | GLM-5 | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 8.1% |
| LMArena Math | 1440 | 1283 |
| MathArena Final-Answer Competitions | 65.7% | — |
| Omni-MATH | — | 33% |
| MATH Level 5 | — | 63.2% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-5 leads
GLM-5: 52.3 (#64), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | GLM-5 | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 87.8% | 49.1% |
| LMArena Expert | 1454 | 1245 |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| Vectara Hallucination Rate | 10.1% | — |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multilingual GLM-5 leads
GLM-5: 53.7 (#58), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | GLM-5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1430 | 1252 |
| LMArena Chinese | 1511 | 1272 |
| LMArena French | 1455 | 1280 |
| LMArena German | 1445 | 1234 |
| LMArena Japanese | 1416 | 1180 |
| LMArena Korean | 1423 | 1188 |
| LMArena Russian | 1436 | 1264 |
| LMArena Spanish | 1454 | 1256 |
Instruction Following GLM-5 leads
GLM-5: 75.2 (#67), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | GLM-5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1428 | 1254 |
| IFEval | — | 80.6% |
Long Context GLM-5 leads
GLM-5: 44.7 (#60), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | GLM-5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1446 | 1282 |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.0 (#38), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | GLM-5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1446 | 1269 |
| LMArena Creative Writing | 1439 | 1221 |
| LMArena Multi-Turn | 1456 | 1272 |
| EQ-Bench Creative Writing | 1601 | — |
| WildBench | — | 80.2% |
Frequently asked questions
Is GLM-5 better than Qwen2.5 72B Instruct?
GLM-5 is the stronger model overall, scoring 46.1 to 31.9 on the Noometry Index.
Which is cheaper, GLM-5 or Qwen2.5 72B Instruct?
GLM-5 is cheaper. It lists at $1 per million input tokens and $3.20 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Is GLM-5 or Qwen2.5 72B Instruct better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 33.2 in the Noometry coding category.
Which has the bigger context window?
GLM-5 does, with 205K tokens against 131K.
How many benchmarks do GLM-5 and Qwen2.5 72B Instruct share?
22 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and Qwen2.5 72B Instruct has 43.