feat(local-models): first-class MLX providers + 1143x local-server ports#56
Merged
Conversation
… 1143x ports Wire mlx_lm.server (`mlx`) and `rapid-mlx` as first-class OpenAI-compatible providers across all four harness presets, and relocate every local model server off the crowded dev/test ports into a contiguous 1143x block beside ollama, so serving a local model never collides with a web dev server: ollama 11434 · llama.cpp 11435 · mlx 11436 · rapid-mlx 11437 · vLLM 11438 - add `mlx` provider (:11436, 6 local MLX models incl. gpt-oss-120b) and staged `rapid-mlx` (:11437, for the brew formula) to pi/opencode/hermes/omp - add packages/local-models/serve/serve-mlx.sh (mirrors serve-llamacpp.sh) - move llama.cpp 8080->11435 and vLLM 8000->11438 in presets, serve scripts (--port), and READMEs; historical docs/eval/* intentionally left at old ports - MLX is the fastest local backend on Apple silicon — worth first-class wiring Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Gni5WQNg9NwTZDszfhYSnf
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Wire MLX as a first-class local backend across all four harness presets, and move every local model server off the crowded dev/test ports into a contiguous 1143x block next to ollama.
mlx_lm.server)Why
mlx_lm.serveris the fastest local backend on Apple silicon (M-series) but wasn't a named provider — gpt-oss-120b et al. were invisible to pi/opencode/hermes/omp.Changes
mlxprovider (:11436, 6 local MLX models) + stagedrapid-mlx(:11437, for the brew formula) to all 4 harness presets.packages/local-models/serve/serve-mlx.sh(mirrorsserve-llamacpp.sh).8080→11435, vLLM8000→11438across presets, serve scripts (--port), and READMEs.docs/eval/*intentionally left at the old ports (they record past runs).Presets validated (JSON + YAML parse; provider blocks resolve). No secrets — localhost endpoints only, per repo convention.
🤖 Generated with Claude Code
Need help on this PR? Tag
@codesmith-botwith what you need. Autofix is disabled.