Model comparison
Deepseek Coder v2 vs GLM-4.7
GLM-4.7 is the stronger model overall, scoring 42.0 to 35.9 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 0 categories and GLM-4.7 in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-4.7 leads 60.9 to 38.2.
Side by side
| Deepseek Coder v2 | GLM-4.7 | |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 35.9 | 42.0 |
| Released | 2024-06-17 | 2025-12-22 |
| Weights | Open | Open |
| Context window | — | 205K |
| Max output | — | 131K |
| Input $ / M tokens | — | $0.60 |
| Output $ / M tokens | — | $2.20 |
| Results tracked | 24 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-4.7 leads
Deepseek Coder v2: 38.1 (#183), GLM-4.7: 44.0 (#79)
| Benchmark | Deepseek Coder v2 | GLM-4.7 |
|---|---|---|
| LMArena Coding | 1251 | 1454 |
| LMArena WebDev | — | 1435 |
| SciCode | — | 45.1% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| ALE-Bench | — | 399.48 |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Agentic & Tool Use Not comparable
Deepseek Coder v2: —, GLM-4.7: 26.5 (#103)
| Benchmark | Deepseek Coder v2 | GLM-4.7 |
|---|---|---|
| Terminal-Bench | — | 33.4% |
| Vending-Bench 2 | — | 2,377 |
Reasoning Too close to call
Deepseek Coder v2: 23.6 (#176), GLM-4.7: 24.3 (#164)
| Benchmark | Deepseek Coder v2 | GLM-4.7 |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1443 |
| SimpleBench | — | 47.7% |
| CritPt | — | 1.7% |
| Chess Puzzles | — | 6% |
| Epoch Capabilities Index | — | 143.51 |
| WinoGrande | 83.7% | — |
Math GLM-4.7 leads
Deepseek Coder v2: 34.9 (#190), GLM-4.7: 38.6 (#135)
| Benchmark | Deepseek Coder v2 | GLM-4.7 |
|---|---|---|
| LMArena Math | 1241 | 1423 |
| OTIS Mock AIME 2024-2025 | — | 83.3% |
| ProofBench | — | 6% |
| FrontierMath (Feb 2025 set) | — | 2.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
| GSM8K | 94.5% | — |
Knowledge GLM-4.7 leads
Deepseek Coder v2: 32.3 (#212), GLM-4.7: 47.0 (#80)
| Benchmark | Deepseek Coder v2 | GLM-4.7 |
|---|---|---|
| LMArena Expert | 1181 | 1424 |
| GPQA Diamond | — | 83.3% |
| SimpleQA Verified | — | 32.2% |
| Vectara Hallucination Rate | — | 11.7% |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual GLM-4.7 leads
Deepseek Coder v2: 36.3 (#240), GLM-4.7: 52.8 (#79)
| Benchmark | Deepseek Coder v2 | GLM-4.7 |
|---|---|---|
| LMArena Non-English | 1182 | 1417 |
| LMArena Chinese | 1201 | 1495 |
| LMArena French | 1185 | 1432 |
| LMArena German | 1164 | 1424 |
| LMArena Japanese | 1126 | 1439 |
| LMArena Korean | 1104 | 1399 |
| LMArena Russian | 1188 | 1423 |
| LMArena Spanish | 1153 | 1434 |
Instruction Following GLM-4.7 leads
Deepseek Coder v2: 61.7 (#242), GLM-4.7: 74.4 (#95)
| Benchmark | Deepseek Coder v2 | GLM-4.7 |
|---|---|---|
| LMArena Instruction Following | 1180 | 1411 |
Long Context GLM-4.7 leads
Deepseek Coder v2: 37.0 (#224), GLM-4.7: 42.8 (#116)
| Benchmark | Deepseek Coder v2 | GLM-4.7 |
|---|---|---|
| LMArena Longer Query | 1219 | 1432 |
| CL-bench | — | 15.9% |
| CL-bench Life | — | 10.9% |
Writing & Preference GLM-4.7 leads
Deepseek Coder v2: 38.2 (#253), GLM-4.7: 60.9 (#93)
| Benchmark | Deepseek Coder v2 | GLM-4.7 |
|---|---|---|
| LMArena Text | 1191 | 1435 |
| LMArena Creative Writing | 1120 | 1401 |
| LMArena Multi-Turn | 1177 | 1446 |
| EQ-Bench Creative Writing | — | 1413 |
Frequently asked questions
Is Deepseek Coder v2 better than GLM-4.7?
GLM-4.7 is the stronger model overall, scoring 42.0 to 35.9 on the Noometry Index.
Is Deepseek Coder v2 or GLM-4.7 better for coding?
GLM-4.7 scores higher on coding benchmarks: 44.0 versus 38.1 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and GLM-4.7 share?
17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and GLM-4.7 has 36.