cMajors commited on
Commit
196a70c
·
verified ·
1 Parent(s): a6150ce

Clarify that the hard tasks declare their own macros

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -67,7 +67,7 @@ A GGUF version for llama.cpp, LM Studio and Ollama is in [maylore-ai/Temper-2-Ax
67
 
68
  ## Limitations
69
 
70
- Axon trails larger models, with Qwen2.5-Coder-7B solving 8.2 points more HumanEval-RS tasks on the first try and Qwen3.5-9B 11.8 more. The model rarely solves tasks with macros, lifetimes or async code. Having been trained on chat requests without fill-in-the-middle, it does not work as a drop-in model for inline completion in an editor.
71
 
72
  ## License
73
 
 
67
 
68
  ## Limitations
69
 
70
+ Axon trails larger models, with Qwen2.5-Coder-7B solving 8.2 points more HumanEval-RS tasks on the first try and Qwen3.5-9B 11.8 more. The model rarely solves tasks that declare their own macros, use lifetimes or async code. Having been trained on chat requests without fill-in-the-middle, it does not work as a drop-in model for inline completion in an editor.
71
 
72
  ## License
73