hypneum-lab/micro-kiki-stack-chat-fr

micro-kiki stack-chat-fr (MLX adapter, French chat)

Single MLX adapter file (adapters.safetensors, 6.76 GB) plus its iteration-1300 checkpoint (0001300_adapters.safetensors, same size). Base model: Qwen/Qwen3.5-35B-A3B (not confirmed by a config file; stated by the authors). Published 2026-05-05 as a "multi-adapter stack for French chat".

Provenance warning. No adapter_config.json, no training log and no eval of this exact file are shipped. Base model, rank, target layers, learning rate and data are therefore not recorded for this artefact; the base model stated above is the one declared by the original card and could not be verified from files. Prefer micro-kiki-v4-sota/chat-fr (fully documented) for any use.

Files and format

Hub revision 4daf8bc5c3ce53c853576b2afbdc4f58e9862fc3 (last modified 2026-10-25T09:09:42Z), 5 files, 13 529 412 476 bytes (13.5 GB): adapters.safetensors, 0001300_adapters.safetensors (MLX adapter format), evals/chat-fr-164-cross.json (see Evaluation), README.md, .gitattributes.

Recipe

Parameter	Value
Base model (declared)	Qwen/Qwen3.5-35B-A3B
Framework	MLX (mlx_lm), BF16
Iterations	≥ 1300 (from the checkpoint file name)
Rank / scale / layers / lr / seq length	not recorded
Training data	micro-kiki chat-fr domain (French instruction/chat corpus, incl. Opus-distilled samples); counts not recorded; the chat-fr corpus as source is an inference, not a documented fact

A same-named, configured adapter exists in hypneum-lab/micro-kiki-v4-sota/chat-fr (Qwen3.6 base, rank 16); the two should not be assumed identical.

Training host & cost

Mac Studio M3 Ultra 512 GB, MLX BF16 (project default). Peak memory, wall-clock, energy: not recorded. Host powered off since 2026-10.

Evaluation

The file evals/chat-fr-164-cross.json kept in this repository reports HumanEval+ (n=164, max_tokens 384) pass@1 = 0.543, total generation 729.7 s — but its own metadata says it was run with base_model = Qwen3.6-35B-A3B and adapter = lora-qwen36-35b-v4-sota/chat-fr, i.e. on the V4-SOTA chat-fr adapter, not on this file. No evaluation of this 6.76 GB adapter is recorded; the file is hosted in this repository but belongs to another adapter.

June 2026 iact-bench caveat: on Qwen3.6-35B-A3B, the sibling ailiance LoRA family showed no gain over the base except html-css; no comparable measurement exists for this adapter. (statement of the 2026-10-25 card; no source file in the repository, unverified on 2026-10-31)

Limitations

Undocumented recipe; research artefact; no safety tuning; not for medical, legal or financial advice. Caveats stated by the source repository: only 3 of 10 probed domains showed measurable improvement over the base model (README.md, section "Stacks vs base"); the project follows an errata policy under which a number is reported only if a committed artefact reproduces it (docs/ERRATA.md). Research artefact of the pre-registered dreamOfkiki programme (parent registration OSF 10.17605/OSF.IO/Q6JYN).

Citation

micro-kiki concept DOI 10.5281/zenodo.19666759.

Provenance

https://github.com/hypneum-lab/micro-kiki · https://github.com/ailiance/ailiance-mac-tuner

• Source repository: https://github.com/hypneum-lab/micro-kiki (Apache-2.0 LICENSE), checked at commit a3968d9e1f0b6a6b8cc4cb028d4fef3ced24770c.
• Zenodo DOI: 10.5281/zenodo.19666759, concept DOI of the source repository; no per-artefact DOI.
• Parent pre-registration: OSF 10.17605/OSF.IO/Q6JYN (CITATION.cff).
• Hub revision described by the file listing: 4daf8bc5c3ce53c853576b2afbdc4f58e9862fc3.
• Card revision date: 2026-10-31.

EU AI Act transparency

LoRA adapter weights only, base unchanged; provider Clément Saillant / L'Electron Rare for Hypneum Lab; limited risk, Article 50; contact clement@saillant.cc. Outputs may be inaccurate, biased or fabricated.

AI assistance (transparency)

This repository is developed by a one-person lab with AI assistants used as tools, never as authors:

• Claude Code (Anthropic) assists with code, documentation, operations and literature search in interactive sessions driven by the PI.
• Codex CLI (OpenAI, gpt-6.1-sol / gpt-6-astra) runs independent review passes before a release, a pre-registration or a submission; the briefs, reports and verdict tables are published next to the reviewed material whenever a pass is run.

Every scientific claim, verdict and pre-registration decision is taken by the PI; results are reproduced by the test suite and CI, and review reports are evidence, not endorsements. No model is listed as a co-author of a publication or as a co-author of a commit.