# DeepSeek-V3.1-Terminus vs GLM-4.6

> DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 41.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v3-1-terminus-vs-glm-4-6
- Last updated: 2026-10-10
- Shared benchmarks: 14

## Summary

- They share 14 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 2 categories and GLM-4.6 in 5 categories; 3 gaps are clear of the uncertainty.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 57.4% for DeepSeek-V3.1-Terminus and 47.4% for GLM-4.6.
- DeepSeek-V3.1-Terminus is cheaper at $0.27 / $1 per million input/output tokens, against $0.60 / $2.20 for GLM-4.6.
- GLM-4.6 accepts more context: 205K tokens versus 164K.

## Snapshot

| | DeepSeek-V3.1-Terminus | GLM-4.6 |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 43.1 | 41.4 |
| Rank | 97 | 135 |
| Context | 164K | 205K |
| Input $/M | $0.27 | $0.60 |
| Output $/M | $1 | $2.20 |
| Weights | Open | Open |

## Coding

- DeepSeek-V3.1-Terminus: 42.0 (#113)
- GLM-4.6: 40.1 (#148)

| Benchmark | DeepSeek-V3.1-Terminus | GLM-4.6 |
|---|---|---|
| SciCode | 40.6% | 38.4% |
| LMArena Coding | 1426 | 1449 |
| ALE-Bench | 745.17 | 340.82 |
| SWE-bench Verified (bash only) | — | 55.4% |
| LMArena WebDev | — | 1340 |

## Agentic & Tool Use

- DeepSeek-V3.1-Terminus: —
- GLM-4.6: 32.3 (#66)

| Benchmark | DeepSeek-V3.1-Terminus | GLM-4.6 |
|---|---|---|
| Terminal-Bench | — | 24.5% |
| Berkeley Function Calling Leaderboard | — | 72.4% |

## Reasoning

- DeepSeek-V3.1-Terminus: 26.4 (#133)
- GLM-4.6: 23.7 (#172)

| Benchmark | DeepSeek-V3.1-Terminus | GLM-4.6 |
|---|---|---|
| Kagi LLM Benchmark | 57.4% | 47.4% |
| CritPt | 1.7% | 1.1% |
| LMArena Hard Prompts | 1426 | 1440 |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |

## Math

- DeepSeek-V3.1-Terminus: 38.5 (#137)
- GLM-4.6: 39.1 (#111)

| Benchmark | DeepSeek-V3.1-Terminus | GLM-4.6 |
|---|---|---|
| LMArena Math | 1402 | 1432 |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 2.1% |

## Knowledge

- DeepSeek-V3.1-Terminus: —
- GLM-4.6: 40.2 (#124)

| Benchmark | DeepSeek-V3.1-Terminus | GLM-4.6 |
|---|---|---|
| Vectara Hallucination Rate | — | 9.5% |
| LMArena Expert | — | 1431 |

## Multilingual

- DeepSeek-V3.1-Terminus: 52.1 (#92)
- GLM-4.6: 53.5 (#66)

| Benchmark | DeepSeek-V3.1-Terminus | GLM-4.6 |
|---|---|---|
| LMArena Non-English | 1407 | 1426 |
| LMArena Russian | 1436 | 1419 |
| LMArena Chinese | — | 1499 |
| LMArena French | — | 1459 |
| LMArena German | — | 1447 |
| LMArena Japanese | — | 1393 |
| LMArena Korean | — | 1400 |
| LMArena Spanish | — | 1436 |

## Instruction Following

- DeepSeek-V3.1-Terminus: 74.0 (#106)
- GLM-4.6: 74.3 (#98)

| Benchmark | DeepSeek-V3.1-Terminus | GLM-4.6 |
|---|---|---|
| LMArena Instruction Following | 1404 | 1410 |

## Long Context

- DeepSeek-V3.1-Terminus: 43.4 (#97)
- GLM-4.6: 43.4 (#94)

| Benchmark | DeepSeek-V3.1-Terminus | GLM-4.6 |
|---|---|---|
| LMArena Longer Query | 1421 | 1422 |

## Writing & Preference

- DeepSeek-V3.1-Terminus: 61.0 (#92)
- GLM-4.6: 61.1 (#90)

| Benchmark | DeepSeek-V3.1-Terminus | GLM-4.6 |
|---|---|---|
| LMArena Text | 1419 | 1440 |
| LMArena Creative Writing | 1403 | 1411 |
| LMArena Multi-Turn | 1411 | 1427 |
| EQ-Bench Creative Writing | — | 1411 |

## FAQ

### Is DeepSeek-V3.1-Terminus better than GLM-4.6?

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 41.4 on the Noometry Index.

### Which is cheaper, DeepSeek-V3.1-Terminus or GLM-4.6?

DeepSeek-V3.1-Terminus is cheaper. It lists at $0.27 per million input tokens and $1 per million output tokens; GLM-4.6 lists at $0.60 and $2.20.

### Is DeepSeek-V3.1-Terminus or GLM-4.6 better for coding?

DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 42.0 versus 40.1 in the Noometry coding category.

### Which has the bigger context window?

GLM-4.6 does, with 205K tokens against 164K.

### How many benchmarks do DeepSeek-V3.1-Terminus and GLM-4.6 share?

14 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and GLM-4.6 has 29.