Skip to main content

One post tagged with "codex"

View All Tags

Day 0 Support: GPT-5.3-Codex

Sameer Kankute
Sameer Kankute
SWE @ LiteLLM (LLM Translation)
Krrish Dholakia
Krrish Dholakia
CEO, LiteLLM
Ishaan Jaff
Ishaan Jaff
CTO, LiteLLM

LiteLLM now supports GPT-5.3-Codex on Day 0, including support for the new assistant phase metadata on Responses API output items.

Why phase matters for GPT-5.3-Codex​

phase appears on assistant output items and helps distinguish preamble/commentary turns from final closeout responses.

Reference: Phase parameter docs

Supported values:

  • null
  • "commentary"
  • "final_answer"

Important:

  • Persist assistant output items with phase exactly as returned.
  • Send those assistant items back on the next turn.
  • Do not add phase to user messages.

Docker Image​

docker pull ghcr.io/berriai/litellm:v1.81.12-stable.gpt-5.3

Usage​

1. Setup config.yaml

model_list:
  - model_name: gpt-5.3-codex
    litellm_params:
      model: openai/gpt-5.3-codex

2. Start the proxy

docker run -d \
  -p 4000:4000 \
  -e ANTHROPIC_API_KEY=$OPENAI_API_KEY \
  -v $(pwd)/config.yaml:/app/config.yaml \
  ghcr.io/berriai/litellm:v1.81.12-stable.gpt-5.3 \
  --config /app/config.yaml

3. Test it

curl -X POST "http://0.0.0.0:4000/v1/responses" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $LITELLM_KEY" \
  -d '{
    "model": "gpt-5.3-codex",
    "input": "Write a Python script that checks if a number is prime."
  }'

Python Example: Persist phase with OpenAI Client + LiteLLM Base URL​

from openai import OpenAI

client = OpenAI(
    base_url="http://0.0.0.0:4000/v1",  # LiteLLM Proxy
    api_key="your-litellm-api-key",
)

items = []  # Persist this per conversation/thread


def _item_get(item, key, default=None):
    if isinstance(item, dict):
        return item.get(key, default)
    return getattr(item, key, default)


def run_turn(user_text: str):
    global items

    # User message: no phase field
    items.append(
        {
            "type": "message",
            "role": "user",
            "content": [{"type": "input_text", "text": user_text}],
        }
    )

    resp = client.responses.create(
        model="gpt-5.3-codex",
        input=items,
    )

    # Persist assistant output items verbatim, including phase
    for out_item in (resp.output or []):
        items.append(out_item)

    # Optional: inspect latest phase for UI/telemetry routing
    latest_phase = None
    for out_item in reversed(resp.output or []):
        if _item_get(out_item, "type") == "output_item.done" and _item_get(out_item, "phase") is not None:
            latest_phase = _item_get(out_item, "phase")
            break

    return resp, latest_phase

Notes​

  • Use /v1/responses for GPT Codex models.
  • Preserve full assistant output history for best multi-turn behavior.
  • If phase metadata is dropped during history reconstruction, output quality can degrade on long-running tasks.