Triangle104/Meissa-Qwen2.5-7B-Instruct-Q8_0-GGUF

This model was converted to GGUF format from Orion-zhen/Meissa-Qwen2.5-7B-Instruct using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model.

---

Model details:

Meissa is designated Lambda Orionis, forms Orion's head, and is a multiple star with a combined apparent magnitude of 3.33. Its name means the "shining one".

This model is fine tuned over writing and role playing datasets (maybe the first on qwen2.5-7b), aiming to enhance model's performance in novel writing and roleplaying.

The model is fine-tuned over Orion-zhen/Qwen2.5-7B-Instruct-Uncensored Training details

I used SFT method. Datasets used are listed below:

anthracite-org/stheno-filtered-v1.1
MinervaAI/Aesir-Preview
Epiculous/Synthstruct-Gens-v1.1-Filtered-n-Cleaned
anthracite-org/nopm_claude_writing_fixed
Gryphe/Sonnet3.5-Charcard-Roleplay
nothingiisreal/DirtyWritingPrompts
Orion-zhen/tagged-pixiv-novel

---

Use with llama.cpp

Install llama.cpp through brew (works on Mac and Linux)

brew install llama.cpp

Invoke the llama.cpp server or the CLI.

CLI:

llama-cli --hf-repo Triangle104/Meissa-Qwen2.5-7B-Instruct-Q8_0-GGUF --hf-file meissa-qwen2.5-7b-instruct-q8_0.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo Triangle104/Meissa-Qwen2.5-7B-Instruct-Q8_0-GGUF --hf-file meissa-qwen2.5-7b-instruct-q8_0.gguf -c 2048

Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.

git clone https://github.com/ggerganov/llama.cpp

Step 2: Move into the llama.cpp folder and build it with LLAMA_CURL=1 flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).

cd llama.cpp && LLAMA_CURL=1 make

Step 3: Run inference through the main binary.

./llama-cli --hf-repo Triangle104/Meissa-Qwen2.5-7B-Instruct-Q8_0-GGUF --hf-file meissa-qwen2.5-7b-instruct-q8_0.gguf -p "The meaning to life and the universe is"

or

./llama-server --hf-repo Triangle104/Meissa-Qwen2.5-7B-Instruct-Q8_0-GGUF --hf-file meissa-qwen2.5-7b-instruct-q8_0.gguf -c 2048