Changelog / release
1.0.4 — Reliable agent launches, image edits and local adapters
Released 2026-11-17 · full changelog · GitHub releases
Rapid-MLX 1.0.4 makes coding-agent launches reliable, adds MiniMax-H3-Image and FLUX 3 Klein, and serves local LoRA adapters without fusing them.
- Coding agents on first launch. Setup configures the agent before waiting for server readiness, with bounded waits and clear messages. Claude Code uses the real context window without a billing notice; removal restores prior settings while keeping later edits. Hermes preserves the first message and cleans up with
--remove. Aider uses the user’s shell and isolated model metadata; Qwen Code gets valid empty settings in its isolated home. (#5804, #5811, #5840, #5813, #5815, #5828, #5818, #5812) - Image generation and multi-reference edits.
minimax-h3-image(4-bit) andminimax-h3-image-8bitsupport up to 16 API reference images; Desktop currently limits attachments to ten. The MiniMax-H3-Image Community License excludes the EU, UK, Republic of Korea and US, and requires separate authorization above US$1M annual revenue or for hosted third-party services. Responses carry attribution; weights download from the upstream Hub only. FLUX 3 Klein and Klein Base support up to ten edit references. Qwen-Image advertises checkpoint-specific limits and returns HTTP 400 when exceeded. (#5832, #5790, #5839) - Image LoRAs. Place your
.safetensorsfiles in~/.rapid-mlx/lorasand select them by name for generation and edits. The image runtime moves to mflux 0.23 with no generation behavior change. (#5792, #5787) - Local text adapters.
rapid-mlx serve <model> --adapter <dir>applies anmlx_lm.loraadapter at startup without fusing or re-quantizing. Base-model checks work with aliases, Hugging Face repo IDs and local cache paths; mismatched bases are rejected. (#5808, #5842) - API and built-in tools.
/v1/completionsreturns prompt-token logprobs withechoandlogprobs. Qwen3-Reranker uses its trained prompt format. Lone UTF-16 surrogates in embedding or rerank input return HTTP 400. The built-in calculator and batch read-only helpers accept native JSON arguments. (#5809, #5843, #5801, #5800) - Cached first-token latency. Single requests hitting long prefix caches avoid extra KV copies; chat prompts are tokenized once per request. Decode speed and outputs are unchanged. (#5831, #5803, #5810, #5829)
- Desktop fixes. Conversation history remains reachable from the compact sidebar rail, and onboarding keeps model names visible beside long size details. (#5799, #5793)
- License detection. The project LICENSE is the verbatim Apache-2.0 text. (#5783)
- Upgrade and Desktop 1.0.4. CLI
v1.0.4and Desktoprapid-mac-v1.0.4were released 2026-11-17 (PT). Runrapid-mlx upgrade,brew upgrade rapid-mlxorpip install -U rapid-mlx==1.0.4. Download Desktop 1.0.4.