← All templates

Ollama

Published, not hardware-testedociLLM inference

Ollama daemon; pull models on demand and call its API through an SSH tunnel.

Needs 7.5 GiB VRAM · CUDA on NVIDIA

No GPU big enough is free right now.

New machines appear as hosts connect them.

No speed benchmarks yet, so we match on price.

Template details
Rating
No ratings yet
Runs
0
Min VRAM
8.1 GB
Backend
CUDA · NVIDIA GPUs
Version
ollama-tunnel-v1.0.0
Image digest
d696622202b0…
Health check
SSH probe on the leased endpoint

SSH as `tenant`; the web UI listens on 127.0.0.1 inside the container only, open it with `ssh -L`. Ollama has no authentication, so its API listens on 127.0.0.1:11434 only: `ssh -L 11434:127.0.0.1:11434`. Models pull into /work and carry their own licences. Thin AmpleRun wrapper (templates/images/workspace, variant ollama-tunnel) over ollama/ollama:0.34.4@sha256:8262851b2846b87c649eddf3e76beb270c52f4d1bc94559f47efde16b0841551 (digest verified 2026-09-26); NOT BUILT, so image_digest is a placeholder that can be neither published nor rented. Licence: MIT (https://github.com/ollama/ollama/blob/main/LICENSE). Status: not hardware-tested.

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

#llm #ollama

Reported on real GPUs

People report these themselves. We haven't checked them.

No reports yet.

Sign in to report how this template runs on your GPU.

Reviews

No reviews yet.

Sign in to leave a review.

Deploy Ollama · AmpleRun