# Deepseek Coder v2 vs GLM-5.1

> GLM-5.1 is the stronger model overall, scoring 47.6 to 35.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-coder-v2-vs-glm-5-1
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 0 categories and GLM-5.1 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GLM-5.1 leads 67.2 to 38.4.

## Snapshot

| | Deepseek Coder v2 | GLM-5.1 |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 35.3 | 47.6 |
| Rank | 217 | 51 |
| Context | — | 200K |
| Input $/M | — | $1.40 |
| Output $/M | — | $4.40 |
| Weights | Open | Open |

## Coding

- Deepseek Coder v2: 37.5 (#178)
- GLM-5.1: 48.5 (#52)

| Benchmark | Deepseek Coder v2 | GLM-5.1 |
|---|---|---|
| LMArena Coding | 1251 | 1486 |
| SWE-bench Verified | — | 74.2% |
| LMArena WebDev | — | 1509 |
| SciCode | — | 43.8% |
| WeirdML | — | 57.1% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| ALE-Bench | — | 887.1 |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |

## Agentic & Tool Use

- Deepseek Coder v2: —
- GLM-5.1: 24.8 (#106)

| Benchmark | Deepseek Coder v2 | GLM-5.1 |
|---|---|---|
| APEX-Agents | — | 40.9% |
| ExploitBench | — | 18.1% |
| GBAEval | — | 0% |
| Vending-Bench 2 | — | 5,634 |

## Reasoning

- Deepseek Coder v2: 22.6 (#175)
- GLM-5.1: 38.2 (#57)

| Benchmark | Deepseek Coder v2 | GLM-5.1 |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1472 |
| SimpleBench | — | 55.1% |
| NYT Connections (extended) | — | 77.7% |
| CritPt | — | 4.6% |
| Chess Puzzles | — | 19% |
| Thematic Generalization | — | 69.8% |
| Epoch Capabilities Index | — | 149.86 |
| WinoGrande | 83.7% | — |

## Math

- Deepseek Coder v2: 32.3 (#188)
- GLM-5.1: 49.0 (#48)

| Benchmark | Deepseek Coder v2 | GLM-5.1 |
|---|---|---|
| LMArena Math | 1241 | 1472 |
| FrontierMath (Tiers 1-3) | — | 36.8% |
| MathArena Final-Answer Competitions | — | 67.1% |
| OTIS Mock AIME 2024-2025 | — | 93.3% |
| ProofBench | — | 22.2% |
| FrontierMath (Feb 2025 set) | — | 33.4% |
| FrontierMath Tier 4 (v1) | — | 12.5% |
| GSM8K | 94.5% | — |

## Knowledge

- Deepseek Coder v2: 31.7 (#209)
- GLM-5.1: 55.0 (#45)

| Benchmark | Deepseek Coder v2 | GLM-5.1 |
|---|---|---|
| LMArena Expert | 1181 | 1477 |
| GPQA Diamond | — | 89.9% |
| SimpleQA Verified | — | 34% |
| ARC (AI2) Challenge | 64.3% | — |

## Multilingual

- Deepseek Coder v2: 36.4 (#235)
- GLM-5.1: 55.2 (#34)

| Benchmark | Deepseek Coder v2 | GLM-5.1 |
|---|---|---|
| LMArena Non-English | 1182 | 1448 |
| LMArena Chinese | 1201 | 1516 |
| LMArena French | 1184 | 1478 |
| LMArena German | 1164 | 1466 |
| LMArena Japanese | 1126 | 1436 |
| LMArena Korean | 1104 | 1423 |
| LMArena Russian | 1188 | 1453 |
| LMArena Spanish | 1153 | 1467 |

## Instruction Following

- Deepseek Coder v2: 61.9 (#237)
- GLM-5.1: 76.5 (#41)

| Benchmark | Deepseek Coder v2 | GLM-5.1 |
|---|---|---|
| LMArena Instruction Following | 1180 | 1451 |

## Long Context

- Deepseek Coder v2: 37.0 (#219)
- GLM-5.1: 44.9 (#51)

| Benchmark | Deepseek Coder v2 | GLM-5.1 |
|---|---|---|
| LMArena Longer Query | 1219 | 1467 |

## Writing & Preference

- Deepseek Coder v2: 38.4 (#249)
- GLM-5.1: 67.2 (#29)

| Benchmark | Deepseek Coder v2 | GLM-5.1 |
|---|---|---|
| LMArena Text | 1191 | 1462 |
| LMArena Creative Writing | 1120 | 1454 |
| LMArena Multi-Turn | 1176 | 1473 |
| EQ-Bench Creative Writing | — | 1592 |

## FAQ

### Is Deepseek Coder v2 better than GLM-5.1?

GLM-5.1 is the stronger model overall, scoring 47.6 to 35.3 on the Noometry Index.

### Is Deepseek Coder v2 or GLM-5.1 better for coding?

GLM-5.1 scores higher on coding benchmarks: 48.5 versus 37.5 in the Noometry coding category.

### How many benchmarks do Deepseek Coder v2 and GLM-5.1 share?

17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and GLM-5.1 has 41.