Skip to content
Draft
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 16 additions & 3 deletions docs/providers/deepseek.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
---
sidebar_label: DeepSeek
description: Configure DeepSeek's chat and reasoning models in Zoo Code. Access deepseek-chat and deepseek-reasoner for coding and complex reasoning tasks.
description: Configure DeepSeek V4 models in Zoo Code through the first-party API and compatible hosted providers.
keywords:
- deepseek
- deepseek chat
Expand All @@ -10,11 +10,12 @@ keywords:
- reasoning ai
- coding ai
- deepseek r1
- deepseek v4 pro 0813
---

# Using DeepSeek With Zoo Code

Zoo Code supports accessing models through the DeepSeek API, including `deepseek-chat` and `deepseek-reasoner`.
Zoo Code supports DeepSeek V4 models through the DeepSeek API and compatible hosted providers. DeepSeek and OpenCode Go route the stable `deepseek-v4-pro` alias to the DeepSeek V4 Pro 0813 checkpoint. Fireworks AI and Baseten offer separate IDs for the published V4 Pro preview weights and the 0813 checkpoint.

**Website:** [https://platform.deepseek.com/](https://platform.deepseek.com/)

Expand All @@ -33,6 +34,17 @@ Zoo Code supports accessing models through the DeepSeek API, including `deepseek

Zoo Code supports all models available through the DeepSeek API.

DeepSeek V4 Pro is available through these Zoo Code providers using each provider's published model ID:

| Provider | Model ID | Version served |
| --- | --- | --- |
| DeepSeek | `deepseek-v4-pro` | 0813 checkpoint |
| OpenCode Go | `deepseek-v4-pro` | 0813 checkpoint |
| Fireworks AI | `accounts/fireworks/models/deepseek-v4-pro` | Published V4 Pro preview weights |
| Fireworks AI | `accounts/fireworks/models/deepseek-v4-pro-0813` | 0813 checkpoint |
| Baseten | `deepseek-ai/DeepSeek-V4-Pro` | Published V4 Pro preview weights |
| Baseten | `deepseek-ai/DeepSeek-V4-Pro-0813` | 0813 checkpoint |

For the complete, up-to-date model list, see [DeepSeek's API documentation](https://api-docs.deepseek.com/quick_start/pricing).

---
Expand All @@ -47,4 +59,5 @@ For the complete, up-to-date model list, see [DeepSeek's API documentation](http
---

## Tips and Notes
* **Pricing:** Refer to the [DeepSeek Pricing](https://api-docs.deepseek.com/quick_start/pricing/) page for details on model costs.
* **First-party pricing:** At 16:00 UTC on August 16, 2026, DeepSeek introduces peak and off-peak pricing. Peak hours are 01:00-04:00 and 06:00-10:00 UTC; all other hours are half-price. Peak V4 Pro rates per million tokens are $1.32 for cache misses, $0.044 for cache hits, and $3.96 for output. Peak V4 Flash rates are $0.44, $0.014, and $1.32 respectively. Zoo Code uses peak rates for static cost estimates.
* **Hosted-provider pricing:** OpenCode Go keeps its published V4 Pro rates of $0.435 input, $0.003625 cached input, and $0.87 output, while V4 Flash remains $0.14, $0.0028, and $0.28. Baseten charges $1.32 input, $0.132 cached input, and $3.96 output for its dated 0813 model; its preview model remains $1.74, $0.145, and $3.48. Fireworks charges $1.32, $0.044, and $3.96 for 0813, while its preview model remains $1.74, $0.145, and $3.48. Provider prices are independent, so refer to each provider's pricing page for future changes.
Loading