# GLM 4.6 Fp8 vs GLM 4.7 FP8

On provider list prices, GLM 4.7 FP8 costs $0.45 per million input tokens against $0.60 for GLM 4.6 Fp8: 1.3x apart. Output is $2 against $2.20 (1.1x). On Allocate both bill at list plus the 7% transaction fee.

## Specifications

| | GLM 4.6 Fp8 | GLM 4.7 FP8 |
| --- | --- | --- |
| Lab | Zai Org | Zai Org |
| Access | Open weights | Open weights |
| Context window | 198K tokens | 198K tokens |
| List price, input | $0.60 / M tokens | $0.45 / M tokens |
| List price, output | $2.20 / M tokens | $2 / M tokens |
| Cached input | n/a | n/a |
| License | MIT | MIT |
| Fine-tunable | Yes | Yes |

Specifications and provider list prices from the Allocate catalog, checked 2026-07-08. Billed price is list plus the 7% transaction fee.

## What the numbers say

Take 1,000,000 requests a month at 1,200 input and 350 output tokens each. That workload costs $1,240 a month on GLM 4.7 FP8 and $1,490 on GLM 4.6 Fp8 at list: a gap of $250, or 1.2x.

## Choose GLM 4.6 Fp8 for

- Fine-tuning under a permissive license (MIT)

## Choose GLM 4.7 FP8 for

- The lower list price ($0.45 in / $2 out per M tokens)
- Fine-tuning under a permissive license (MIT)

## Common questions

### Which is cheaper, GLM 4.6 Fp8 or GLM 4.7 FP8?

GLM 4.7 FP8, on this workload shape. At list prices it is $0.45/$2 per million tokens in and out against $0.60/$2.20 for GLM 4.6 Fp8. Billed on Allocate: $0.48/$2.14 against $0.64/$2.35, list plus 7%.

### Which has the bigger context window?

They match: both read 202,752 tokens (198K) per request.

### Can I fine-tune GLM 4.6 Fp8 or GLM 4.7 FP8?

Both publish open weights (GLM 4.6 Fp8: MIT; GLM 4.7 FP8: MIT), so both can be fine-tuned. On Allocate the trained weights stay inside your boundary and belong to you.

---

[HTML page](https://allocate.network/compare/z-ai-glm-4-6-vs-z-ai-glm-4-7) · [GLM 4.6 Fp8](https://allocate.network/models/z-ai-glm-4-6.md) · [GLM 4.7 FP8](https://allocate.network/models/z-ai-glm-4-7.md) · [Machine-readable catalog](https://allocate.network/catalog.json)