NeelM0906 / Mference
Swift + Metal MoE inference for Apple Silicon: Qwen 3.6 35B at 23.5–29.3 tok/s decode with 2.20× faster long-prompt prefill on a 24 GB M5; Gemma 4 26B in ~2 GB, DeepSeek-V4-Flash 284B, Inkling-Small 276B, native Mac app, CLI, and OpenAI-compatible server.
Why this job: topics: apple-silicon, llm-inference
What it does
A native Swift + Metal runtime for pinned checkpoints, with bounded-memory streaming and measured ports.
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Active
- Last commit 6 days ago
- Stars
- 122
- On GitHub
- Licence
- Custom
- Read before business use
- Downloads for
- None listed
- No release files
- Latest version
- v0.1.0
- Released 50 days ago
- Built with
- Swift
- 85% of the code
Is it looked after?
- Contributors
- Not stated
- Last 90 days
- Issues
- 0% closed
- 1 still open
- Pull requests
- 43 merged
- 3 waiting
- Releases
- 1
- 1 in the last year
- Contributing guide (opens in new tab)
- Security policy (opens in new tab)
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| incoai / splash | 2,170 | 4 days ago | Apache-2.0 | Mac |
| drumih / turbo-fieldfare | 7,326 | 15 days ago | Apache-2.0 | No downloads |
| carloslfu / slotstream | 662 | yesterday | MIT | No downloads |
More: languages, topics and project record
Languages
- Swift 85%
- Metal 8%
- Python 5%
- Jinja 1%
Topics
- apple-silicon
- deepseek
- gemma
- llm-inference
- macos
- metal
- mixture-of-experts
- qwen
- swift
Project record
- JobLocal LLMs
- StatusListed
- Commits553
- Forks18
- Watching5
- Stars in 28 days+1
- CreatedJul 31, 2026
- First seen by usNov 7, 2026
- CheckedNov 8, 2026