upload eu-kiki LoRA adapter

README.md

@@ -0,0 +1,90 @@
+---
+license: apache-2.0
+base_model: swiss-ai/Apertus-70B-Instruct-2509
+tags:
+  - lora
+  - peft
+  - mlx
+  - eu-kiki
+  - eu-ai-act
+language:
+  - fr
+  - en
+library_name: peft
+---
+
+# eu-kiki-apertus-math-lora
+
+LoRA adapter for **swiss-ai/Apertus-70B-Instruct-2509**, part of the [eu-kiki](https://github.com/L-electron-Rare/eu-kiki) project — a 100 % EU-sovereign multi-model LLM serving pipeline. EU AI Act Article 52/53 compliant.
+
+## Performance
+
+⚠️ Trained but **not yet evaluated on a public math benchmark** (training loss converged, validation loss = 0.5xx). Use at your own risk; we recommend pairing with a verifier.
+
+## Usage
+
+```python
+from mlx_lm import load
+from mlx_lm.tuner.utils import linear_to_lora_layers
+from huggingface_hub import snapshot_download
+
+base_path = snapshot_download("swiss-ai/Apertus-70B-Instruct-2509")
+adapter_path = snapshot_download("clemsail/eu-kiki-apertus-math-lora")
+
+model, tokenizer = load(base_path)
+linear_to_lora_layers(model, num_layers=32, config={"rank": 16, "alpha": 32})
+model.load_weights(f"{adapter_path}/adapters.safetensors", strict=False)
+```
+
+Or, simpler, fuse and serve via `mlx_lm fuse`:
+
+```bash
+python -m mlx_lm fuse \
+    --model swiss-ai/Apertus-70B-Instruct-2509 \
+    --adapter-path <adapter_path> \
+    --save-path /tmp/eu-kiki-apertus-math-lora-fused \
+    --dequantize
+```
+
+## Training configuration
+
+| Parameter | Value |
+|---|---|
+| Method | LoRA |
+| Rank | 16 |
+| Alpha | 32 |
+| Dropout | 0.05 |
+| Target modules | q_proj, k_proj, v_proj, o_proj |
+| Precision | BF16 |
+| Optimiser | AdamW |
+| Learning rate | 1e-5 |
+| Framework | MLX (`mlx_lm` fork on Apple Silicon) |
+| Hardware | Mac Studio M3 Ultra 512 GB unified memory |
+
+## Provenance & EU AI Act compliance
+
+Datasets used to train this adapter are HF-traceable. Per-source SPDX licenses, download dates, source row counts, and used row counts are documented in:
+
+- [`docs/eu-ai-act-transparency.md`](https://github.com/L-electron-Rare/eu-kiki/blob/main/docs/eu-ai-act-transparency.md) — system-level transparency record (Art. 52/53)
+- [`MODEL_CARD.md`](https://github.com/L-electron-Rare/eu-kiki/blob/main/MODEL_CARD.md) — full evaluation summary across HumanEval+, MT-Bench, GSM8K, KIKI-DSL v3
+- [`eval/results/SUMMARY.md`](https://github.com/L-electron-Rare/eu-kiki/blob/main/eval/results/SUMMARY.md) — per-bench reproducible results
+
+## Risk classification
+
+**Limited risk** (EU AI Act Art. 52). General-purpose AI; not deployed in safety-critical contexts.
+
+## License
+
+Apache 2.0, matching the base model.
+
+## Citation
+
+```bibtex
+@misc{eu-kiki-2026,
+  title  = {eu-kiki: EU-sovereign multi-model LLM serving with HF-traceable LoRA adapters},
+  author = {Saillant, Clément},
+  year   = {2026},
+  url    = {https://github.com/L-electron-Rare/eu-kiki},
+  note   = {Live demo: https://ml.saillant.cc}
+}
+```

adapter_config.json

@@ -0,0 +1,16 @@
+{
+  "fine_tune_type": "lora",
+  "lora_parameters": {
+    "rank": 16,
+    "alpha": 32,
+    "dropout": 0.05,
+    "scale": 2.0
+  },
+  "num_layers": 16,
+  "lora_layers": [
+    "self_attn.q_proj",
+    "self_attn.k_proj",
+    "self_attn.v_proj",
+    "self_attn.o_proj"
+  ]
+}

adapters.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:982d5a5de7713b56a036dfc44df1d337d32ef279e2f3b89dccc470a5c1748936
+size 786538979