sleepy/llama.cpp
1
0
Fork 0
Code Issues 8 Pull Requests 1 Actions 175 Packages Projects Releases Wiki Activity

model : add tokenizer from LFM2.5-Audio-1.5B (#19687)

Browse Source 
* model : Add tokenizer from LFM2.5-Audio-1.5B

[LFM2.5-Audio-1.5B](https://huggingface.co/LiquidAI/LFM2.5-Audio-1.5B) introduced lightweight audio tokenizer.

Tokenizer based on LFM2 architecture and acts as "embedding" model with
different input `n_embd` and output `n_embd_out`.

To be used in https://github.com/ggml-org/llama.cpp/pull/18641.

To convert use

```shell
python3 convert_hf_to_gguf.py /path/to/LFM2.5-Audio-1.5B/audio_detokenizer
```

* Update convert_hf_to_gguf.py

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* Formatting

* Rework check for attention layers

* Add LFM2 SWA model support

* Address PR feedback

* Set vocab to none

* Move helper function definitions to cpp file

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
Tarek Dakhran
2026-02-19 09:54:48 +01:00
committed by GitHub
parent eacb4b67a2
commit 8004f3a8d1
7 changed files with 185 additions and 137 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/models/models.h
+1 -7
View File
@@ -347,15 +347,9 @@ struct llm_build_kimi_linear : public llm_build_delta_net_base {
const llama_model & model;
};
template <bool iswa>
struct llm_build_lfm2 : public llm_graph_context {
const llama_model & model;
llm_build_lfm2(const llama_model & model, const llm_graph_params & params);
ggml_tensor * build_moe_feed_forward(ggml_tensor * cur, int il) const;
ggml_tensor * build_dense_feed_forward(ggml_tensor * cur, int il) const;
ggml_tensor * build_attn_block(ggml_tensor * cur, ggml_tensor * inp_pos, llm_graph_input_attn_kv * inp_attn, int il) const;
ggml_tensor * build_shortconv_block(ggml_tensor * cur, llm_graph_input_rs * inp_recr, int il);
};
struct llm_build_llada : public llm_graph_context {