Upload folder using huggingface_hub

.gitattributes

@@ -49,3 +49,4 @@ gemma-4-E2B-it-int4-aggr-v5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-4-E2B-it-int4-aggr-v2/gemma-4-E2B-it-int4-aggr-v2.cellmd filter=lfs diff=lfs merge=lfs -text
 gemma-4-E2B-it-int4-aggr-v2/gemma-4-E2B-it-int4-aggr-v2.tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Bonsai-1.7B_v2/Bonsai-1.7B_v2.cellm filter=lfs diff=lfs merge=lfs -text
+lfm2.5-350m-v1/lfm2.5-350m-v1.cellm filter=lfs diff=lfs merge=lfs -text

lfm2.5-350m-v1/README.md

@@ -0,0 +1,35 @@
+# LFM2.5-350M (cellm)
+
+Liquid Foundation Model 2.5 (350M) converted to cellm format. Pre-quantized to 4-bit using MLX.
+
+## Model details
+
+- **Base model**: liquid-ai/LFM2.5-350M-MLX-4bit
+- **Format**: cellm (.cellm)
+- **Precision**: 4-bit weights (MLX int4), f16 norms
+- **Size**: 211 MB
+- **Architecture**: 16 layers, hybrid LIV convolution + grouped-query attention
+  - Pattern: conv, conv, attn, conv, conv, attn, conv, conv, attn, conv, attn, conv, attn, conv, attn, conv
+- **Hidden size**: 1024
+- **Vocab**: 65,536
+
+## Usage
+
+```sh
+./target/release/infer \
+  --model lfm2.5-350m-v1.cellm \
+  --tokenizer tokenizer.json \
+  --prompt "The capital of France is" \
+  --gen 40 --temperature 0 --backend cpu --kv-encoding f16
+```
+
+## Notes
+
+- Text completion only. This is a base model.
+- 350M parameters at 4-bit gives 211 MB on disk.
+- The LIV (Linear Input-Varying) conv blocks use a causal sliding-window state per layer.
+- Use temperature > 0 for more varied output.
+
+## License
+
+Apache 2.0 (same as upstream LFM2.5)

lfm2.5-350m-v1/lfm2.5-350m-v1.cellm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca9dd89946b22fafea82d85091df2c8b7d0f8d5fa2280154d78b29235f322350
+size 221688000