Model comparison
Deepseek Coder v2 vs GLM-4.5-Air
GLM-4.5-Air is the stronger model overall, scoring 38.7 to 35.3 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeekSummary
- They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 1 category and GLM-4.5-Air in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-4.5-Air leads 56.1 to 38.4.
Side by side
| Deepseek Coder v2 | GLM-4.5-Air | |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 35.3 | 38.7 |
| Released | 2024-06-17 | 2025-07-20 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 98K |
| Input $ / M tokens | — | $0.20 |
| Output $ / M tokens | — | $1.10 |
| Results tracked | 24 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Deepseek Coder v2 leads
Deepseek Coder v2: 37.5 (#178), GLM-4.5-Air: 34.5 (#225)
| Benchmark | Deepseek Coder v2 | GLM-4.5-Air |
|---|---|---|
| LMArena Coding | 1251 | 1397 |
| GSO | — | 2.9% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning Too close to call
Deepseek Coder v2: 22.6 (#175), GLM-4.5-Air: 23.3 (#166)
| Benchmark | Deepseek Coder v2 | GLM-4.5-Air |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1379 |
| Kagi LLM Benchmark | — | 43% |
| ForecastBench | — | 59.2 |
| WinoGrande | 83.7% | — |
Math GLM-4.5-Air leads
Deepseek Coder v2: 32.3 (#188), GLM-4.5-Air: 33.8 (#166)
| Benchmark | Deepseek Coder v2 | GLM-4.5-Air |
|---|---|---|
| LMArena Math | 1241 | 1396 |
| Omni-MATH | — | 39.1% |
| GSM8K | 94.5% | — |
Knowledge GLM-4.5-Air leads
Deepseek Coder v2: 31.7 (#209), GLM-4.5-Air: 34.6 (#184)
| Benchmark | Deepseek Coder v2 | GLM-4.5-Air |
|---|---|---|
| LMArena Expert | 1181 | 1370 |
| Humanity's Last Exam | — | 8.1% |
| MMLU-Pro | — | 76.2% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 59.4% |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual GLM-4.5-Air leads
Deepseek Coder v2: 36.4 (#235), GLM-4.5-Air: 49.4 (#130)
| Benchmark | Deepseek Coder v2 | GLM-4.5-Air |
|---|---|---|
| LMArena Non-English | 1182 | 1366 |
| LMArena Chinese | 1201 | 1427 |
| LMArena French | 1184 | 1399 |
| LMArena German | 1164 | 1376 |
| LMArena Japanese | 1126 | 1348 |
| LMArena Korean | 1104 | 1306 |
| LMArena Russian | 1188 | 1373 |
| LMArena Spanish | 1153 | 1385 |
Instruction Following GLM-4.5-Air leads
Deepseek Coder v2: 61.9 (#237), GLM-4.5-Air: 69.8 (#165)
| Benchmark | Deepseek Coder v2 | GLM-4.5-Air |
|---|---|---|
| LMArena Instruction Following | 1180 | 1353 |
| IFEval | — | 81.2% |
Long Context GLM-4.5-Air leads
Deepseek Coder v2: 37.0 (#219), GLM-4.5-Air: 41.6 (#130)
| Benchmark | Deepseek Coder v2 | GLM-4.5-Air |
|---|---|---|
| LMArena Longer Query | 1219 | 1367 |
Writing & Preference GLM-4.5-Air leads
Deepseek Coder v2: 38.4 (#249), GLM-4.5-Air: 56.1 (#134)
| Benchmark | Deepseek Coder v2 | GLM-4.5-Air |
|---|---|---|
| LMArena Text | 1191 | 1384 |
| LMArena Creative Writing | 1120 | 1343 |
| LMArena Multi-Turn | 1176 | 1371 |
| WildBench | — | 78.9% |
Frequently asked questions
Is Deepseek Coder v2 better than GLM-4.5-Air?
GLM-4.5-Air is the stronger model overall, scoring 38.7 to 35.3 on the Noometry Index.
Is Deepseek Coder v2 or GLM-4.5-Air better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 37.5 versus 34.5 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and GLM-4.5-Air share?
17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and GLM-4.5-Air has 27.