Decision 3.0 vs Granite Switch 4.1

Decision 3.0

Open Apache-2.0 decision models that read images and video as well as text and answer several typed questions in one call, with a probability per option and tagged Hub releases. They are two days old, need a GPU and trust_remote_code, drop Vela's set and span types, and publish no calibration figures.

Granite Switch 4.1

One self-hosted Apache 2.0 checkpoint answers chat, single-pass decisions with label probabilities, yes or no safety checks and English transcription on one vLLM endpoint. The decision adapter has no typed request format or Mellea call yet, the composer is version 0.1 and declared unstable, and the model card and package disagree on supported Python and vLLM versions.

Decision 3.0

1. Pin a tag or commit with revision= when loading from the Hub. Each model had five tagged releases on 10 and 11 October
2. Send images and videos to d3_server.py as base64 data URLs only. The server refuses paths and http URLs with 422
3. Set DECISION_API_KEY before binding the bundled server beyond 127.0.0.1. It is open otherwise
4. Turn on permutation_average=True when routing must not depend on option order. It is off by default
5. Expect one request at a time from the bundled server. Use the router's model runtime for queued serving

Granite Switch 4.1

1. Pass chat_template_kwargs: {"adapter_name": "decision"} or "guardian-core". Without it the request is ordinary chat on the base model
2. For a decision, send the state and the ### DECISION question as two user turns, request logprob_token_ids for the option labels, and set max_completion_tokens to 1
3. Use single-token option labels. Split more than 128 labels into chunks with the identical prompt and apply one softmax over all returned logprobs
4. Parse guardian-core answers as JSON, {"score": "yes"} or {"score": "no"}, where yes means the criterion is met
5. Add authentication before exposing the vLLM server. Neither vLLM nor the checkpoint requires a key by default