# DeepSeek vs Z.ai

> DeepSeek and Z.ai both ship MIT-licensed weights from China. DeepSeek wins on per-token price and off-peak discounts; Z.ai wins on a flat-rate coding plan and drop-in Claude Code support.

Canonical: https://www.subconscious.dev/compare/deepseek-vs-z-ai · By The Subconscious Team · Updated September 30, 2026

## How they compare

On paper these labs look alike: Chinese, open weights under MIT, cheap hosted APIs. The pricing models differ. DeepSeek bills per token with a time-of-day split, so V4 Pro is $1.32 in and $3.96 out at peak and half that off-peak. Z.ai's GLM-5.3 is $1.40 in and $4.40 out with cached input at $0.26, GLM-5.3-Flash is $0.075 in and $0.25 out, and several older Flash models are free. DeepSeek's cache hits cost a few cents per million or less, and its models carry 1M context with 384K max output.

Z.ai's pull is the GLM Coding Plan. At $18 a month on Lite, developers get a quota Z.ai says equals 15 to 30x the fee at API rates, and an Anthropic-compatible endpoint runs Claude Code on GLM with a few environment variables. The catch is that quota burns 2 to 3x faster on premium models during Beijing peak hours. DeepSeek's catch is that its hosted data is stored in China and it reprices often. Individual coders on a budget lean Z.ai. Metered agents and batch jobs lean DeepSeek.

## What each one does

### DeepSeek

DeepSeek is the Chinese lab whose open-weight models reset price expectations for the whole market. Its API now serves two models, both with 1M context and 384K max output. V4.1 Flash shipped September 10, 2026 with built-in image understanding at $0.30 in and $1.20 out at peak. V4 Pro, generally available since August 13, costs $1.32 in and $3.96 out at peak. Cache hits cost a few cents per million or less, and the weights ship on Hugging Face under an MIT license.

### Z.ai

Z.AI is the international brand of Chinese lab Zhipu AI, maker of the GLM models. Its current flagship, GLM-5.3, shipped August 17, 2026 at $1.40 in and $4.40 out per million tokens, with cached input at $0.26. GLM-5.3-Flash costs $0.075 in and $0.25 out, and several older Flash models are priced at zero, a real free tier instead of trial credits. GLM-5, released in February 2026, is a 744B mixture-of-experts model under an MIT license, and at launch it ranked first among open-weight models on the Artificial Analysis index with a record-low hallucination score.

## Which is best, and when

### Choose DeepSeek for

- Metered agents and batch work priced per token
- Long outputs up to 384K tokens
- Very cheap cache hits on long prefixes

### Choose Z.ai for

- Flat-rate agentic coding inside Claude Code
- Free Flash models for prototyping
- Developers who want predictable monthly spend

## At a glance

| Attribute | DeepSeek | Z.ai |
|---|---|---|
| Model access | Open weights (MIT) | Open weights (MIT) |
| Flagship models | DeepSeek V4.1 Flash, V4 Pro | GLM-5.3, GLM-5.3-Flash |
| Speed | ~35 tok/s on V4 Pro | ~80 tok/s on GLM-5.3 |
| Price | Off-peak hours at half price | $1.40 in, $4.40 out (GLM-5.3); free Flash tier |
| Customization | Open weights to fine-tune | Open weights, no license limits |
| Deployment | First-party API, Hugging Face weights | API, GLM Coding Plan |
| Long context | 1M, 384K max output | 1M (GLM-5.3) |

## FAQ

### What is the difference between DeepSeek and Z.ai?

DeepSeek and Z.ai both ship MIT-licensed weights from China. DeepSeek wins on per-token price and off-peak discounts; Z.ai wins on a flat-rate coding plan and drop-in Claude Code support.

### When should I choose DeepSeek over Z.ai?

Metered agents and batch work priced per token; Long outputs up to 384K tokens; Very cheap cache hits on long prefixes.

### When should I choose Z.ai over DeepSeek?

Flat-rate agentic coding inside Claude Code; Free Flash models for prototyping; Developers who want predictable monthly spend.

### Is DeepSeek or Z.ai cheaper?

DeepSeek: Off-peak hours at half price. Z.ai: $1.40 in, $4.40 out (GLM-5.3); free Flash tier. The cheaper choice depends on the model and workload.

### Which has more context, DeepSeek or Z.ai?

DeepSeek: 1M, 384K max output. Z.ai: 1M (GLM-5.3).

## Running long-horizon agents?

If your agents run past 200K tokens, compare both against Subconscious: [Subconscious vs DeepSeek](https://www.subconscious.dev/compare/subconscious-vs-deepseek.md), [Subconscious vs Z.ai](https://www.subconscious.dev/compare/subconscious-vs-z-ai.md).

Full profiles: [DeepSeek](https://www.subconscious.dev/providers/deepseek.md), [Z.ai](https://www.subconscious.dev/providers/z-ai.md).