tokspur
Private chat and code on your Mac. Run a 177B model on a 128 GB M5 Max: local Qwen3.8 Flash-Next inference with a 65,536-token context, Tokspur's own Metal engine with native MTP, and an OpenAI-compatible API. Apache 2.0.
Get picks like this daily. Today's launch numbers, the top launches, AI and tech news, and a daily opportunity pick, straight to your inbox.
Want to reach these readers? Sponsor an issue.
GITHUB · launched Nov 9, 2026 · 13 upvotes (top 96%) · 0 views
Niche: local ai model runners and clients · 110 launches, 6 in the last 7 days ▼ -57% · Explore this niche →
Category and tags
- Horizontal
- Model & infra
- Prosumer
- AI-native
- Commercial product
- apple-silicon
- gguf
- inference-engine
- llm
- llm-inference
- local-inference
- +11 more
Visit website ↗ Source listing ↗
More like this: all 110 launches in local ai model runners and clients →
Could you build this?
No The product is a custom Metal inference engine that runs a 177B model with speculative decoding, which requires GPU kernel and systems expertise beyond what vibe coding produces.
What it would actually take: A real version needs a custom Metal compute engine with hand-tuned kernels for quantised matmul and attention, memory management to fit a large model in unified memory, and an implementation of multi-token prediction. It also needs model conversion tooling and benchmarking against llama.cpp. This takes experienced systems and ML-kernel engineers, not a vibe-coded app wrapper.
Competitors
Other products that read as similar to this one : 142 launches clear the similarity bar, closest 8 shown.
Attention rank: #139 of 143 (itself plus its competitors, highest first, normalized so YC and Product Hunt are compared fairly).
Launched 342 days after the earliest competitor.
- Running 104GB Qwen3.8-Flash-Next on 48GB Mac with at ~12 tok/s · hn · 2026-09-01 · 240 upvotes · similarity 0.51
- molten · github · 2026-10-25 · 22 upvotes · similarity 0.48
- Samosa Chat · hn · 2026-07-15 · 6 upvotes · similarity 0.48
- Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Mac · hn · 2026-07-29 · 919 upvotes · similarity 0.46
- collabosm · github · 2026-09-25 · 90 upvotes · similarity 0.46
- Run 500B+ Parameter LLMs Locally on a Mac Mini · hn · 2026-03-09 · 17 upvotes · similarity 0.45
- 816B-Parameter AI from Your Laptop · ph · 2026-10-23 · 1 upvotes · similarity 0.43
- Fastest Qwen 3.8 27 on single RTX5090 · ph · 2026-09-07 · 1 upvotes · similarity 0.43
The niche, labels, vibe code rating and competitor comparison on this page are analysis by Nicheloom. If you use them, please credit nicheloom.com. How you may use our data.