# GLM-4.5 vs GLM-4.6

> GLM-4.5 and GLM-4.6 score almost the same on the Noometry Index (42.0 vs 41.4), so choose on price, context window or the category you care about most.

- Canonical page: https://noometry.com/compare/glm-4-5-vs-glm-4-6
- Last updated: 2026-10-11
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. GLM-4.5 scores higher in 2 categories and GLM-4.6 in 6 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in long context, where GLM-4.6 leads 43.4 to 38.2.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 57.9% for GLM-4.5 and 47.4% for GLM-4.6.
- Both cost about the same: $0.60 input and $2.20 output per million tokens.
- GLM-4.6 accepts more context: 205K tokens versus 131K.

## Snapshot

| | GLM-4.5 | GLM-4.6 |
|---|---|---|
| Provider | Z.ai (Zhipu) | Z.ai (Zhipu) |
| Noometry Index | 42.0 | 41.4 |
| Rank | 122 | 135 |
| Context | 131K | 205K |
| Input $/M | $0.60 | $0.60 |
| Output $/M | $2.20 | $2.20 |
| Weights | Open | Open |

## Coding

- GLM-4.5: 41.4 (#125)
- GLM-4.6: 40.1 (#148)

| Benchmark | GLM-4.5 | GLM-4.6 |
|---|---|---|
| SWE-bench Verified (bash only) | 54.2% | 55.4% |
| LMArena Coding | 1434 | 1449 |
| ALE-Bench | 344.82 | 340.82 |
| LMArena WebDev | — | 1340 |
| SciCode | — | 38.4% |
| WeirdML | 40.6% | — |
| AlgoTune | 1.52 | — |

## Agentic & Tool Use

- GLM-4.5: —
- GLM-4.6: 32.3 (#66)

| Benchmark | GLM-4.5 | GLM-4.6 |
|---|---|---|
| Terminal-Bench | — | 24.5% |
| Berkeley Function Calling Leaderboard | — | 72.4% |

## Reasoning

- GLM-4.5: 28.6 (#100)
- GLM-4.6: 23.7 (#172)

| Benchmark | GLM-4.5 | GLM-4.6 |
|---|---|---|
| Kagi LLM Benchmark | 57.9% | 47.4% |
| LMArena Hard Prompts | 1429 | 1440 |
| CritPt | — | 1.1% |

## Math

- GLM-4.5: 39.0 (#116)
- GLM-4.6: 39.1 (#111)

| Benchmark | GLM-4.5 | GLM-4.6 |
|---|---|---|
| LMArena Math | 1427 | 1432 |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 2.1% |

## Knowledge

- GLM-4.5: 35.9 (#179)
- GLM-4.6: 40.2 (#124)

| Benchmark | GLM-4.5 | GLM-4.6 |
|---|---|---|
| LMArena Expert | 1433 | 1431 |
| Humanity's Last Exam | 8.3% | — |
| Confabulations | 11.3% | — |
| Vectara Hallucination Rate | — | 9.5% |

## Multilingual

- GLM-4.5: 52.8 (#77)
- GLM-4.6: 53.5 (#66)

| Benchmark | GLM-4.5 | GLM-4.6 |
|---|---|---|
| LMArena Non-English | 1417 | 1426 |
| LMArena Chinese | 1465 | 1499 |
| LMArena French | 1418 | 1459 |
| LMArena German | 1407 | 1447 |
| LMArena Japanese | 1415 | 1393 |
| LMArena Korean | 1380 | 1400 |
| LMArena Russian | 1414 | 1419 |
| LMArena Spanish | 1454 | 1436 |

## Instruction Following

- GLM-4.5: 74.1 (#104)
- GLM-4.6: 74.3 (#98)

| Benchmark | GLM-4.5 | GLM-4.6 |
|---|---|---|
| LMArena Instruction Following | 1404 | 1410 |

## Long Context

- GLM-4.5: 38.2 (#201)
- GLM-4.6: 43.4 (#94)

| Benchmark | GLM-4.5 | GLM-4.6 |
|---|---|---|
| LMArena Longer Query | 1412 | 1422 |
| Fiction.LiveBench | 58.3% | — |

## Writing & Preference

- GLM-4.5: 57.5 (#127)
- GLM-4.6: 61.1 (#90)

| Benchmark | GLM-4.5 | GLM-4.6 |
|---|---|---|
| LMArena Text | 1430 | 1440 |
| LMArena Creative Writing | 1395 | 1411 |
| EQ-Bench Creative Writing | 1343 | 1411 |
| LMArena Multi-Turn | 1415 | 1427 |
| Short-Story Creative Writing | 73.4% | — |

## FAQ

### Is GLM-4.5 better than GLM-4.6?

GLM-4.5 and GLM-4.6 score almost the same on the Noometry Index (42.0 vs 41.4), so choose on price, context window or the category you care about most.

### Which is cheaper, GLM-4.5 or GLM-4.6?

GLM-4.6 is cheaper. It lists at $0.60 per million input tokens and $2.20 per million output tokens; GLM-4.5 lists at $0.60 and $2.20.

### Is GLM-4.5 or GLM-4.6 better for coding?

GLM-4.5 scores higher on coding benchmarks: 41.4 versus 40.1 in the Noometry coding category.

### Which has the bigger context window?

GLM-4.6 does, with 205K tokens against 131K.

### How many benchmarks do GLM-4.5 and GLM-4.6 share?

21 benchmarks have published results for both models. GLM-4.5 has 27 scored results on Noometry and GLM-4.6 has 29.