vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

model: Add support for PhiMoE arch (#11003)

Browse source 
* model: support phimoe

* python linter

* doc: minor

Co-authored-by: ThiloteE <73715071+ThiloteE@users.noreply.github.com>

* doc: minor

Co-authored-by: ThiloteE <73715071+ThiloteE@users.noreply.github.com>

* doc: add phimoe as supported model

ggml-ci

---------

Co-authored-by: ThiloteE <73715071+ThiloteE@users.noreply.github.com>
This commit is contained in:
Pierrick Hymbert 2025-01-09 11:21:41 +01:00 • committed by GitHub
parent be0e950c91
commit f8feb4b01a
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
10 changed files with 208 additions and 31 deletions
Download patch file Download diff file Expand all files Collapse all files

10
docs/development/HOWTO-add-model.md
View file

@ -28,7 +28,7 @@ The required steps to implement for an HF model are:
```python
@Model.register("MyModelForCausalLM")
class MyModel(Model):
model_arch = gguf.MODEL_ARCH.GROK
model_arch = gguf.MODEL_ARCH.MYMODEL
```
2. Define the layout of the GGUF tensors in [constants.py](/gguf-py/gguf/constants.py)
@ -79,14 +79,14 @@ Depending on the model configuration, tokenizer, code and tensors layout, you wi
- `Model#set_vocab`
- `Model#write_tensors`
NOTE: Tensor names must end with `.weight` suffix, that is the convention and several tools like `quantize` expect this to proceed the weights.
NOTE: Tensor names must end with `.weight` or `.bias` suffixes, that is the convention and several tools like `quantize` expect this to proceed the weights.
### 2. Define the model architecture in `llama.cpp`
The model params and tensors layout must be defined in `llama.cpp`:
1. Define a new `llm_arch`
2. Define the tensors layout in `LLM_TENSOR_NAMES`
3. Add any non standard metadata in `llm_load_hparams`
3. Add any non-standard metadata in `llm_load_hparams`
4. Create the tensors for inference in `llm_load_tensors`
5. If the model has a RoPE operation, add the rope type in `llama_rope_type`
@ -96,9 +96,9 @@ NOTE: The dimensions in `ggml` are typically in the reverse order of the `pytorc
This is the funniest part, you have to provide the inference graph implementation of the new model architecture in `llama_build_graph`.
Have a look at existing implementation like `build_llama`, `build_dbrx` or `build_bert`.
Have a look at existing implementations like `build_llama`, `build_dbrx` or `build_bert`.
When implementing a new graph, please note that the underlying `ggml` backends might not support them all, support for missing backend operations can be added in another PR.
Some `ggml` backends do not support all operations. Backend implementations can be added in a separate PR.
Note: to debug the inference graph: you can use [llama-eval-callback](/examples/eval-callback/).