DevDock-AI / Speedy
Qwen3.8-Flash-Next (125B MoE) up to 1M context on one or two RTX cards (tuned on 3080 10gb). A measured, tuned and benchmarked fork of Strata.
Why this job: topics: llama-cpp, inference, llama
What it does
Each value is the average ± half the spread across the measured runs. All runs use medium reasoning and real prompts.
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Active
- Last commit 3 days ago
- Stars
- 56
- +31 this week
- Licence
- MIT
- Allows business use
- Downloads for
- 2 systems
- Windows, Linux
- Latest version
- v0.3.1
- Released 4 days ago
- Built with
- C++
- 44% of the code
Is it looked after?
- Contributors
- 8
- Last 90 days
- Issues
- None
- No issues filed
- Pull requests
- 3 merged
- 0 waiting
- Releases
- 2
- 2 in the last year
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| vllm-project / vllm | 96,588 | today | Apache-2.0 | Mac, Linux |
| wici-ai / inferssd | 364 | today | Apache-2.0 | Mac, Windows |
| SemiAnalysisAI / InferenceX | 2,010 | today | Apache-2.0 | No downloads |
More: download files, languages, topics and project record
Download files, v0.3.1
- Windows4 files, 186 downloads
- Linux4 files, 65 downloads
Languages
- C++ 44%
- Python 28%
- Cuda 12%
- JavaScript 8%
Topics
- cuda
- inference
- llm
- long-context
- mixture-of-experts
- multi-gpu
- qwen
- rtx-3080
- llama
- llama-cpp
- nvidia
- rtx
Project record
- JobLocal LLMs
- StatusNew to our list
- Commits75
- Forks1
- Watching0
- CreatedNov 1, 2026
- First seen by usNov 7, 2026
- CheckedNov 9, 2026