Granite Switch 4.1 vs Kev

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.

Kev

Apache-2.0 code, adapters and heads on Apache-2.0 Qwen bases, with release tarballs and SHA-256 checksums for the 0.8B, 4B and 9B models. No package. pip install kev installs an unrelated 2021 ORM, so Kev runs from a Git clone with uv.

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

Kev

1. Install from the repository. The kev package on PyPI is an unrelated project
2. Pin a checkpoint with @v1.0, as in jaredpalmer/kev-4b@v1.0, so tuned thresholds keep their meaning
3. Keep states under 8,192 tokens on Kev-0.8B, 4B and 9B, or use Kev-27B for long documents
4. Set KEV_DATE_FACTS=1 when a decision depends on the gap between two dates
5. Expect a 422 naming the token count when a state passes 65,536 tokens. The server refuses it instead of cutting it