6Robert123 / cmp100-100-unlock
Unlock NVIDIA CMP 100-100 mining card from 6 GB to full 12 GB HBM2 on Linux. ~$25 GP100 card for local LLMs (Qwen3.8-27B ~32 tok/s on two cards).
Why this job: topics: llama-cpp, local-llm, llm-inference, local llms
What it does
Hi, I'm 6Robert123 👋 I picked up an NVIDIA CMP 100-100 (GP100 / Pascal) second-hand for about $25. It's a mining card that ships with 6 GiB enabled, and I wanted to see how much more it could do. After a lot of digging it now comes up on Linux with 12,288 MiB reported by the NVIDIA driver. A two-card hot --apply trial passed on 2 November 2026, and the historical evidence runs through 1 November 2026.
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Active
- Last commit 8 days ago
- Stars
- 12
- +3 this week
- Licence
- MIT
- Allows business use
- Downloads for
- None listed
- No release files
- Latest version
- v2026.11.02.2
- Released 10 days ago
- Built with
- C
- 56% of the code
Is it looked after?
- Contributors
- 1
- Last 90 days
- Issues
- 0% closed
- 2 still open
- Pull requests
- 0 merged
- 0 waiting
- Releases
- 1
- 1 in the last year
- Contributing guide (opens in new tab)
- No security policy
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| nicorobinreal / llmspeed | 173 | 23 days ago | MIT | No downloads |
| L-Forster / open-jet | 141 | 2 days ago | AGPL-3.0 | Mac, Windows, Linux |
| Freim32 / vllmops | 101 | 41 days ago | Apache-2.0 | No downloads |
More: languages, topics and project record
Languages
- C 56%
- Python 24%
- Shell 17%
- JavaScript 3%
Topics
- gp100
- gpu
- hbm2
- linux
- llama-cpp
- llm-inference
- local-llm
- mining-gpu
- nvidia
- pascal
- cmp-100-100
- cmp100-100
Project record
- JobLocal LLMs
- StatusNew to our list
- Commits3
- Forks1
- Watching0
- CreatedNov 1, 2026
- First seen by usNov 7, 2026
- CheckedNov 12, 2026