README.md
1.6 KB · 63 lines · markdown Raw
1 ---
2 tags:
3 - gguf
4 - llama.cpp
5 - Anthropic Claude Fable 5
6 - Evol-Instruct-Code-80k-v1
7 - vision-language-model
8 - vllm
9 - ollama
10 - text-generation
11 - code
12 - unsloth
13 license: ecl-2.0
14 datasets:
15 - nickrosh/Evol-Instruct-Code-80k-v1
16 - Qwen/DeepPlanning
17 metrics:
18 - code_eval
19 base_model:
20 - nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4
21 - Qwen/Qwen3.5-9B
22 new_version: Tesleum/shirdel-coder-9b-claude-fable-5
23 pipeline_tag: text-generation
24 library_name: transformers
25 ---
26 [<img src="https://huggingface.co/Tesleum/shirdel-coder-9b-claude-fable-5/resolve/main/shirdel-coder-fable-5.webp" width="600"/>](https://tesleum.com)
27
28 # shirdel-coder-9b-claude-fable-5 : GGUF
29
30 Claude Fable 5–style instruction tuning combined with Evol-Instruct-Code-80k-v1.
31 Fine-tuned and converted to GGUF format using Q6_K quantization.
32
33 **Example usage**:
34 - For text only LLMs: `llama-cli -hf Tesleum/shirdel-coder-9b-claude-fable-5 --jinja`
35 - For multimodal models: `llama-mtmd-cli -hf Tesleum/shirdel-coder-9b-claude-fable-5 --jinja`
36 [![Website](https://upload.wikimedia.org/wikipedia/commons/3/31/Ollama-logo.svg)](https://ollama.com/Tesleum/shirdel-coder)
37
38 # vLLM
39
40 Install:
41
42 ```bash
43 pip install vllm
44 ```
45
46 Serve:
47
48 ```bash
49 vllm serve "Tesleum/shirdel-coder-9b-claude-fable-5"
50 ```
51
52 Or specify tokenizer explicitly:
53
54 ```bash
55 vllm serve "Tesleum/shirdel-coder-9b-claude-fable-5" \
56 --tokenizer Qwen/Qwen3.5-9B
57 ```
58
59
60 ## Available Model files:
61 - `Shirdel-Coder-9B-Claude-Fable-5.Q6_K.gguf`
62 - `Shirdel-Coder-9B-Claude-Fable-5.BF16-mmproj.gguf`
63 [<img src="https://avatars.githubusercontent.com/u/63446410?v=4" width="200"/>](https://tesleum.com)