File size: 410 Bytes
cab2adc
 
fbd237e
 
cab2adc
01359d3
cab2adc
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
---
hub:
  repo_id: umarigan/LLama-3-8B-Instruction-tr-Q4_K_M-GGUF
  filename: llama-3-8b-instruction-tr.Q4_K_M.gguf
llama_cpp:
  n_ctx: 4096
  # n_gpu_layers: 40  # llama 13b has 40 layers
chat:
  stop:
    - "</s>"
    - "<unk>"
    - "### USER:"
    - "USER:"
queue:
  max_size: 16
  concurrency_count: 1  # leave this at 1, llama-cpp-python doesn't handle concurrent requests and will crash the entire app