Models
The model Kourier.sh currently serves behind its compatible API endpoints.
kourier.sh currently exposes one model to customer API keys:
| Model | API model string | Maximum context length | Notes |
|---|---|---|---|
| DeepSeek V4.1 Flash | deepseek-v4.1-flash | 262,144 tokens (256K) | Live coding and reasoning model. |
Choosing the model
Use the exact identifier deepseek-v4.1-flash in OpenAI Chat Completions,
OpenAI Responses, Codex, and both Claude Code model variables. Bifrost does
not translate aliases such as deepseek-v4-flash; an unregistered model
identifier is rejected.
Migrating from earlier model ids
DeepSeek V4.1 Flash replaced DeepSeek V4 Flash on 2026-10-01. The earlier
model strings DSV4-Flash-0731 and DSV4.1-Flash are retired and return
HTTP 400. Your API key, base URL, and plan are unchanged — existing keys were
updated automatically — so the only change is the model string:
| Where it's set | Change it to |
|---|---|
OpenAI SDK / curl model field | deepseek-v4.1-flash |
Codex ~/.codex/config.toml → model | deepseek-v4.1-flash |
Claude Code ANTHROPIC_MODEL and ANTHROPIC_SMALL_FAST_MODEL | deepseek-v4.1-flash |
Oh My Pi models.yml model id and kourier/... roles | deepseek-v4.1-flash / kourier/deepseek-v4.1-flash |
GET https://api.kourier.sh/v1/models returns the exact string your key is
granted. See the changelog
for everything that changed.
Context limits
DeepSeek V4.1 Flash has a maximum context length of 262,144 tokens (256K),
counting the prompt and generated output together. With the 16,384-token
per-response output cap, size the prompt so that the prompt plus requested
output remain within the 262,144-token total.
Routing
The model currently routes through Kourier's internal
kourier-dsv4-autoround provider. Provider names are diagnostic metadata,
not client configuration: clients send the model string, base URL, and their
Kourier API key.
Access
New accounts start on Free with dashboard access. Subscribe to Starter or Pro in Billing to create an API key. Starter includes one concurrent request for $20/month; Pro includes three for $50/month, with no request-volume or token-usage quotas.