Skip to content
Availablegpt-oss

gpt-oss-lite API

21B total parameters, fp4. Fast and inexpensive, with the same tool calling and structured output support as its larger sibling.

Function callingReasoningStreamingStructured Output
Context
131K tokens
Params
21B
Quant
fp4

Call it with any OpenAI client

model: gpt-oss-lite
terminal
curl https://api.kalla.website/v1/chat/completions \
  -H "Authorization: Bearer $KALLA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-oss-lite",
    "messages": [{ "role": "user", "content": "Hello" }],
    "stream": true
  }'

Already using the OpenAI SDK? Change the base URL to https://api.kalla.website/v1 and pass gpt-oss-lite as the model.

OpenAI-compatible

Same paths, same JSON, same SSE frames as OpenAI — existing SDKs and editors work unchanged.

Streaming by default

Read tokens as they are generated; end the stream with the standard [DONE] sentinel.

Tool calling & JSON

Function calling, structured output and multi-turn tool loops for agentic workflows.

Fair-use limits

Per-key rate limits and monthly caps keep the shared fleet responsive at peak load.

Other models on the same key

Switch per request — no new integration.

Available
01

gpt-oss-pro

Open-weight 120B mixture-of-experts with strong reasoning per euro.

Plan
2,500 IQD/ month
Use it
idgpt-oss-pro
AvailableNew
02

Qwen3.8

Vision-capable 27B with strong reasoning and tool use.

Plan
2,500 IQD/ month
Use it
idQwen3.8
AvailableNew
03

Qwen3.5

Flagship mixture-of-experts for the hardest reasoning.

Plan
2,500 IQD/ month
Use it
idQwen3.5