m10dis
Qwen3.8-Flash-Next-FP8 (48-layer MoE, 512 experts) on one RTX 3070 Ti plus 24 Maxwell GPUs from six Tesla M10 cards. PCIe multicast, GPU page-table remapping, 96 tok/s decode.
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 4, 2026 · 36 upvotes (top 73%) · 0 views
Niche: llm inference engines · 133 launches, 4 in the last 7 days ▼ -33% · Explore this niche →
Category and tags
- Horizontal
- Model & infra
- Developer
- AI-native
- Hobby / open-source project
Visit website ↗ Source listing ↗
More like this: all 133 launches in llm inference engines →
Could you build this?
No Running a 48-layer MoE model across a mix of consumer and legacy datacenter GPUs with PCIe multicast and GPU page-table remapping is low-level systems and CUDA research. Vibe coding cannot supply the driver-level and kernel-performance expertise this requires.
What it would actually take: A real version needs custom CUDA kernels, inference runtime work, and experiments on PCIe peer-to-peer and multicast behavior across mixed Maxwell and Ampere hardware. It also needs GPU memory management below the normal driver API, plus a lot of profiling to reach the stated decode speed. This takes engineers with GPU systems and inference-engine experience, not only AI-assisted coding.
Competitors
Other products that read as similar to this one : 211 launches clear the similarity bar, closest 8 shown.
Attention rank: #161 of 212 (itself plus its competitors, highest first, normalized so YC and Product Hunt are compared fairly).
Launched 365 days after the earliest competitor.
- Speedy · github · 2026-11-01 · 35 upvotes · similarity 0.59
- glyd-condor · github · 2026-10-14 · 29 upvotes · similarity 0.56
- qwen38-flash-next-w4a16-cmp170hx · github · 2026-09-16 · 10 upvotes · similarity 0.52
- colab-qwen38-g4 · github · 2026-11-10 · 50 upvotes · similarity 0.52
- Running 104GB Qwen3.8-Flash-Next on 48GB Mac with at ~12 tok/s · hn · 2026-09-01 · 240 upvotes · similarity 0.51
- collabosm · github · 2026-09-25 · 90 upvotes · similarity 0.49
- Strata · github · 2026-09-24 · 918 upvotes · similarity 0.49
- Fastest Qwen 3.8 27 on single RTX5090 · ph · 2026-09-07 · 1 upvotes · similarity 0.48
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.