Qwen2.5-Coder 32B

llama.cpp · Q4_K_M · 20 GB

Qwen2.5-Coder 32B for editors and coding agents, 4-bit GGUF on llama.cpp; fits 24 GB cards; runs on NVIDIA, AMD, Intel Arc and Apple silicon.

Chat in your browser, plus an OpenAI-compatible API.

Publishedllamacppgguf-q4_k_mLLM inference

Needs 22 GiB of GPU memory · Metal on Apple silicon

Pick a machine

CheapestBalancedFastest

Max

Picked Apple M4 Pro (20-core GPU): the lowest price that fits. No speed data for these GPUs yet.

2 machines can run this now1× Apple M4 Pro (20-core GPU)Recommended$0.115/h48 GB unified · 36 GB for AI · 🇷🇴 Romania · Dedicated · New hostNo spec data, matched on price1× Apple M5 Max (40-core GPU)$0.258/h48 GB unified · 37 GB for AI · 🇺🇸 United States · Dedicated · New hostNo spec data, matched on price

Customize

Sort by

GPU maker

GPU model

GPUs

VRAM per GPU

Region

Reliability

Availability

Disk

Verified hosts only

Runs 1 h, then stops · holds $0.115 · incl. 10% fee

Deploying accepts the rental terms. Unused budget comes back to your balance.

Options

Duration limit1 h

Duration in hoursHard cap: the machine stops at this limit even if you forget. Snaps to 30-minute steps up to a day, whole days after.

Budget cap

Held when it starts, never exceeded. Unused budget comes back.

Add SSH public keyOptional. You chat in the browser without one. Add it to also connect with SSH.

Doesn't fit (2)

1× Apple M4 · New hostinsufficient VRAM: need 23622320128 bytes, device has 114525470721× Apple M1 Pro (14-core GPU) · New hostinsufficient VRAM: need 23622320128 bytes, device has 12712935424

Template details

Rating: No ratings yet
Runs: 0
Min VRAM: 24 GB
Backend: Metal on Apple silicon
Weights: 20 GB
Version: llamacpp-qwen2-5-coder-32b-v2-v2.0.0
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 #code #openai-api #gguf

Reviews

No reviews yet.