Huihui QwQ 32B abliterated

QwQ 32B with refusals removed (abliterated), 4-bit GGUF. Chat in your browser, plus an OpenAI-compatible API.

Engine: llama.cpp · Q4_K_M
Size: 20 GB
GPU memory: 22 GiB or more
Runs on

Pick a machine

No machine can run this right now.

New machines appear as hosts connect them.

Have a GPU? Earn with it See all machines

Doesn't fit (3)

1× Apple M5 Max (40-core GPU) · New hostthis host has opted out of uncensored models1× Apple M3 Max (30-core GPU) · New hostthis host has opted out of uncensored models1× Apple M4 Pro (20-core GPU) · New hostthis host has opted out of uncensored models

Hardware that can run this

Nothing that can run this is online right now.

11 more devices could run it, none online

Hardware	Fits	Est. speed
M4 Pro · 24 GB unified	Q4_K_M	8–12 tok/s
RTX 3090 · 24 GB	Q4_K_M	28–41 tok/s
RTX 4090 · 24 GB	Q4_K_M	30–44 tok/s
RX 7900 XTX · 24 GB	Q4_K_M	Evidence unavailable: no cited memory bandwidth for this hardware
M1 Max · 32 GB unified	Q4_K_M	12–18 tok/s
M2 Max · 32 GB unified	Q4_K_M	12–18 tok/s
M3 Max · 48 GB unified	Q4_K_M, Q8	9–13 tok/s
M4 Max · 48 GB unified	Q4_K_M, Q8	12–18 tok/s
M4 Pro · 48 GB unified	Q4_K_M, Q8	8–12 tok/s
Instinct MI210 · 64 GB	Q4_K_M, Q8	Evidence unavailable: no cited memory bandwidth for this hardware
M4 Max · 64 GB unified	Q4_K_M, Q8	12–18 tok/s

Speeds are estimates from memory bandwidth and weight size, not measured benchmarks. A price is the cheapest listed offer for that hardware.

Template details

Rating: No ratings yet
Runs: 0
Min VRAM: 24 GB
Backend: Set by the machine you pick
Weights: 20 GB
Version: llamacpp-huihui-qwq-32b-abliterated-v1.0.0
Image digest: 1f4b9cf58982…
Health check: SSH probe on the leased endpoint

We match GPUs on their measured runtime and dedicated VRAM. A hardware report doesn't qualify another vendor's backend.

#llm #chat #openai-api #gguf #uncensored #abliterated #reasoning

Reviews

No reviews yet.