# DeepSeek V3.2 API

DeepSeek V3.2 is DeepSeek's reasoning-first model built for agents, released on 1 December 2025 and served on Seedrise with a 164K-token context window and up to 8K output tokens; list price $0.64 in / $1.91 out per 1M tokens, provider list as at 1 November 2026.

- Provider: DeepSeek
- Model ID: deepseek-v3.2
- Type: Chat
- Region: AU (can be pinned to Sydney or Melbourne on request)
- Context window: 164,000 tokens
- Max output: 8,000 tokens
- Modalities: text, tools, reasoning
- Endpoints: OpenAI Chat Completions https://api.seedrise.ai/v1
- Prices checked: 1 November 2026

## Provider list price

Provider list price as at 1 November 2026, per 1M tokens. AUD at a reference rate of 1 USD = 1.53 AUD; invoices use the rate on the day.

| Token type | USD | AUD |
|---|---:|---:|
| Input | 0.64 | 0.98 |
| Output | 1.91 | 2.92 |
| Cached input | Rates on request | Rates on request |

Your Seedrise rate is quoted per account against this list price. Request a rate card: https://seedrise.ai/contact?model=deepseek-v3.2

Pinning to an Australian region carries the provider's regional premium on top of the list price shown.

Marked AU: can be pinned to Sydney or Melbourne on request. The Australian-region deployment has an end-of-life date not before 1 December 2026; we will give written notice and suggest a replacement before it is withdrawn.

## Strengths

- Can be pinned to Sydney or Melbourne on request
- DeepSeek's first model to integrate thinking directly into tool use
- Tool use in both thinking and non-thinking modes, per DeepSeek
- 164K-token context window for long documents

## Typical use cases

- Australian workloads that need a DeepSeek model with in-region inference
- Agents and tool-calling pipelines with data residency requirements
- Document Q&A within a 164K-token window
- Cost-sensitive reasoning tasks with short outputs

## Quickstart

A first request with curl. Replace sk-... with your Seedrise key.

```bash
curl https://api.seedrise.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-..." \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-v3.2","messages":[{"role":"user","content":"Hello"}]}'
```

OpenAI SDK, Python:

```python
from openai import OpenAI

client = OpenAI(base_url="https://api.seedrise.ai/v1", api_key="sk-...")

r = client.chat.completions.create(
    model="deepseek-v3.2",
    messages=[{"role": "user", "content": "Summarise this contract in three bullets."}],
)
print(r.choices[0].message.content)
```

## Where it runs

AU: Can be pinned to Sydney or Melbourne on request. Pinning is set per key, not per request.

## Related models

- [DeepSeek V4 Pro](https://seedrise.ai/models/deepseek-v4-pro): US$1.32 input · US$3.96 output per 1M tokens, Global
- [DeepSeek V4.1 Flash](https://seedrise.ai/models/deepseek-flash): US$0.30 input · US$1.20 output per 1M tokens, Global
- [Kimi K2.5](https://seedrise.ai/models/kimi-k2.5): US$0.62 input · US$3.09 output per 1M tokens, AU
- [Mistral Large 4](https://seedrise.ai/models/mistral-large-4): US$0.70 input · US$2.10 output per 1M tokens, Global
- [Gemini 3.8 Flash](https://seedrise.ai/models/gemini-3.8-flash): US$0.75 input · US$3.75 output per 1M tokens, Global

## FAQ

### What is the model ID for DeepSeek V3.2?

Use deepseek-v3.2 as the model field, with no vendor prefix, on the OpenAI-compatible endpoint at https://api.seedrise.ai/v1. GET /v1/models returns the IDs your key can call.

### How does the Seedrise rate for DeepSeek V3.2 compare to the list price?

DeepSeek's list price as at 1 November 2026 is US$0.64 per 1M input tokens and US$1.91 per 1M output tokens. Seedrise quotes rates per account against the provider's published list price for the same routing. Volume and committed tiers sit below list. Australian-region routing carries the provider's regional premium.

### What is the context window and max output of DeepSeek V3.2?

DeepSeek V3.2 has a context window of 164,000 tokens and a maximum output of 8,000 tokens, per the provider documentation checked on 1 November 2026.

### Can DeepSeek V3.2 run in Australia?

Yes. DeepSeek V3.2 can be pinned to Sydney or Melbourne on request. Pinning is set per key, not per request, and carries the provider's regional premium.

### How long will DeepSeek V3.2 stay available?

Its end-of-life date is not before 1 December 2026, and no withdrawal date had been set as at 1 November 2026. We give written notice and a suggested replacement before any model leaves the rate card.

### Why is the output limit on DeepSeek V3.2 8K tokens?

8K output tokens is the limit of the Australian-region deployment, alongside a 164K-token context window. For longer outputs on global routing, DeepSeek V4 Pro and V4.1 Flash allow up to 384K output tokens.

Model names are trademarks of their respective providers.

Source: https://seedrise.ai/models/deepseek-v3.2