How to use from
llama.cpp
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf skywolf46/ignition-singularity-prototype
# Run inference directly in the terminal:
llama cli -hf skywolf46/ignition-singularity-prototype
Install from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf skywolf46/ignition-singularity-prototype
# Run inference directly in the terminal:
llama cli -hf skywolf46/ignition-singularity-prototype
Use pre-built binary
# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf skywolf46/ignition-singularity-prototype
# Run inference directly in the terminal:
./llama-cli -hf skywolf46/ignition-singularity-prototype
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf skywolf46/ignition-singularity-prototype
# Run inference directly in the terminal:
./build/bin/llama-cli -hf skywolf46/ignition-singularity-prototype
Use Docker
docker model run hf.co/skywolf46/ignition-singularity-prototype
Quick Links

Ignition Singularity

λ°œν™” 특이점 ν”„λ‘œμ νŠΈλŠ” AI의 μ»¨ν…μŠ€νŠΈ 보쑰λ₯Ό μœ„ν•΄ μ΅œλŒ€ν•œ μ›λ³Έμ˜ 감정과 λŠλ‚Œμ„ λ˜μ‚΄λ¦΄ 수 μžˆλ„λ‘ ν”„λ‘¬ν”„νŠΈλ₯Ό 압좕을 μ‹œλ„ν•˜λŠ” μƒμ„±ν˜• λͺ¨λΈ ν”„λ‘œμ νŠΈμž…λ‹ˆλ‹€.

ν•™μŠ΅ 정보

λ°œν™” 특이점 ν”„λ‘œμ νŠΈλŠ” 6GB κ°€λŸ‰μ˜ ShareGPT와 CC0, ν˜Ήμ€ μ €μž‘κΆŒμ΄ 만료된 κ³ μ „ μ†Œμ„€ 데이터λ₯Ό Qwen 3.8 27B 및 Gemma 4 31Bλ₯Ό 톡해 μ •μ œν•˜κ³ , λΆ„λ₯˜ν•œ 데이터셋을 톡해 ν•™μŠ΅λ˜μ—ˆμŠ΅λ‹ˆλ‹€.

이 μ •λ³΄λŠ” 변경될 수 있으며, νŒ€μ—μ„œ κ³„μ†ν•˜μ—¬ 쑰건을 λ°”κΏ” ν•™μŠ΅μ„ μ‹œλ„ν•˜κ³  μžˆμŠ΅λ‹ˆλ‹€.

ν•΄λ‹Ή λ ˆν¬μ§€ν† λ¦¬λŠ” μž‘λ™ν•˜λŠ” ν”„λ‘œν† νƒ€μž…λ§Œμ΄ κ³΅κ°œλ˜μ–΄ μžˆμŠ΅λ‹ˆλ‹€.

μ‚¬μš© 방법

λ°œν™” 특이점 ν”„λ‘œν† νƒ€μž… λͺ¨λΈμ€ Qwen 3.5 4B λͺ¨λΈμ„ 기반으둜 파인 νŠœλ‹ν•˜μ˜€μŠ΅λ‹ˆλ‹€.

μ„±λŠ₯이 크게 μ’‹μ§€λŠ” μ•ŠμœΌλ‚˜, μ •μ œλœ κ³ μ „ μ†Œμ„€μ„ ν†΅ν•œ 파인 νŠœλ‹μ„ μ§„ν–‰ν•œ μž‘μ€ λͺ¨λΈμ— λŒ€ν•΄ ꡐ차 검증, ν˜Ήμ€ μ‹€ν—˜μ΄ ν•„μš”ν•œ 경우 λ‚˜μ˜μ§€ μ•Šμ€ 선택지가 될 수 μžˆμŠ΅λ‹ˆλ‹€.

이 ν”„λ‘œμ νŠΈλŠ” IGX νŒ€μ—μ„œ μ§„ν–‰ν•˜λŠ” λ°œν™” 특이점 ν”„λ‘œμ νŠΈμ˜ μΌλΆ€μž…λ‹ˆλ‹€.

Downloads last month
21
GGUF
Model size
4B params
Architecture
qwen35
Hardware compatibility
Log In to add your hardware

We're not able to determine the quantization variants.

Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for skywolf46/ignition-singularity-prototype

Finetuned
Qwen/Qwen3.5-4B
Quantized
(394)
this model