MegEngine / MegEngine/InferLLM

没有支持最新的llama.cpp的格式吗

Open
#38 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
752
Forks
94
PR merge metrics
No merged PRs in 30d

Description

我使用llama.cpp量化过的alpaca7b_plus,看样子并没有支持
```
llama_model_load_internal: format = ggjt v3 (latest)
```

```
main: seed = 1686927440
Assert ' 0 ' failed at file : /mnt/e/pyCode/InferLLM/src/graph/llama.cpp
line 37 : virtual void inferllm::LlamaGraph::load(std::shared_ptr, inferllm::LlmParams&, std::shared_ptr),
extra message: unsupported model type.Aborted
```

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start at graph/llama.cpp line 37, where loading the quantized alpaca7b_plus model asserts on the unsupported model type. Compare the logged ggjt v3 format with the model types handled by this loader, then reproduce the failure with the reported model. Done means the latest llama.cpp-quantized format loads without the unsupported-model assertion.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.