{"slug": "fastflowlm", "name": "FastFlowLM", "category": "ai", "category_name": "AI tools and apps", "description": "", "recommenders": 5, "commenters": 8, "warners": 1, "score": 4.129, "rising": -0.31, "first": "2026-01-09", "latest": "2026-10-07", "why": "a transformer framework called FastFlowLM [0] supporting XDNA 2 NPUs officially started supporting Linux", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 5, "neutral": 3, "compare": 0, "warn": 1}, "spark": [0, 0, 0, 0, 2, 1, 0, 1, 0, 0, 0, 1], "listed": true, "loved": [], "caveat": "", "short_quote": "a transformer framework called FastFlowLM [0] supporting XDNA 2 NPUs officially started supporting Linux", "kind": "", "icon": true, "domain": null, "github_repo": "github.com/fastflowlm/fastflowlm", "official_url": "https://github.com/FastFlowLM/FastFlowLM", "pricing": "", "aliases": ["FastFlowLM"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 799919296, "item_id": 49994956, "url": "https://news.ycombinator.com/item?id=49994956", "by": "Infernal", "date": "2026-10-07", "version": null, "stance": "recommend", "quote": "FastFlowLM is what you're looking for", "excerpt": "FastFlowLM is what you're looking for - Lemonade does package it, as sibling comment says. https://fastflowlm.com/", "context": "", "use_case": "NPU model inference", "switched_from": "", "story": "Strands Decider 2B: a small, open-source, decision model"}, {"id": 776515794, "item_id": 48532237, "url": "https://news.ycombinator.com/item?id=48532237", "by": "jinnko", "date": "2026-06-14", "version": null, "stance": "recommend", "quote": "Check out FastFlowLM for AMD NPU support.", "excerpt": "Check out FastFlowLM for AMD NPU support.", "context": "", "use_case": "AMD NPU inference", "switched_from": "", "story": "Linux 7.1"}, {"id": 761845248, "item_id": 47615328, "url": "https://news.ycombinator.com/item?id=47615328", "by": "cpburns2009", "date": "2026-04-02", "version": null, "stance": "recommend", "quote": "The NPU works on Linux (Arch at least) on Strix Halo using FastFlowLM", "excerpt": "The NPU works on Linux (Arch at least) on Strix Halo using FastFlowLM [1]. Their NPU kernels are proprietary though (free up to a reasonable amount of commercial revenue). It's neat you can run some models basically for free (using NPU instead of CPU/GPU), but the performance is underwhelming. The target for NPUs is really low power devices, and not useful if you have an APU/GPU like Strix Halo. [1]: https://github.c", "context": "", "use_case": "", "switched_from": "", "story": "Lemonade by AMD: a fast and open source local LLM server using GPU and NPU"}, {"id": 756219840, "item_id": 47263740, "url": "https://news.ycombinator.com/item?id=47263740", "by": "jakogut", "date": "2026-03-05", "version": null, "stance": "recommend", "quote": "a transformer framework called FastFlowLM [0] supporting XDNA 2 NPUs officially started supporting Linux", "excerpt": "You in fact can now! In the past week, a transformer framework called FastFlowLM [0] supporting XDNA 2 NPUs officially started supporting Linux. I posted it here the same day I found and started using it, to almost no reaction. [0] https://github.com/FastFlowLM https://fastflowlm.com/ https://huggingface.co/FastFlowLM", "context": "", "use_case": "running models on AMD NPUs", "switched_from": "", "story": "AMD will bring its \"Ryzen AI\" processors to standard desktop PCs for first time"}, {"id": 755653664, "item_id": 47228354, "url": "https://news.ycombinator.com/item?id=47228354", "by": "pstuart", "date": "2026-03-02", "version": null, "stance": "recommend", "quote": "Finally there's good news for Linux users", "excerpt": "Finally there's good news for Linux users: https://github.com/FastFlowLM/FastFlowLM", "context": "", "use_case": "running LLMs on AMD NPU", "switched_from": "", "story": "Welcome (back) to Macintosh"}]}], "caveats": [{"label": "", "summary": "", "quotes": [{"id": 761822112, "item_id": 47613882, "url": "https://news.ycombinator.com/item?id=47613882", "by": "zozbot234", "date": "2026-04-02", "version": null, "stance": "warn", "quote": "NPU-accelerated kernels are proprietary binaries", "excerpt": "That won't give you NPU support, which relies on https://github.com/FastFlowLM/FastFlowLM . And that says \"NPU-accelerated kernels are proprietary binaries\", not open source.", "context": "", "use_case": "", "switched_from": "", "story": "Lemonade by AMD: a fast and open source local LLM server using GPU and NPU"}]}], "who": [], "tags": [], "use_cases": ["npu model inference", "amd npu inference", "running models on amd npus", "running llms on amd npu"], "compared": [{"slug": "lemonade", "name": "Lemonade", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=47612724", "title": "Lemonade by AMD: a fast and open source local LLM server using GPU and NPU", "mentions": 3}, {"url": "https://news.ycombinator.com/item?id=49987076", "title": "Strands Decider 2B: a small, open-source, decision model", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48528729", "title": "Linux 7.1", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=47216825", "title": "AMD will bring its \"Ryzen AI\" processors to standard desktop PCs for first time", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=47224118", "title": "Welcome (back) to Macintosh", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=47171798", "title": "FastFlowLM (FLM) \u2013 Unlock Ryzen AI NPUs", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {"neutral": 1}, "2026-02": {"neutral": 1}, "2026-03": {"recommend": 2}, "2026-04": {"warn": 1, "neutral": 1, "recommend": 1}, "2026-05": {}, "2026-06": {"recommend": 1}, "2026-07": {}, "2026-08": {}, "2026-09": {}, "2026-10": {"recommend": 1}}, "versions": [], "version_unnamed": 0, "family_ref": null, "runs_on": null, "lines": [], "quotes": [{"id": 799919296, "item_id": 49994956, "url": "https://news.ycombinator.com/item?id=49994956", "by": "Infernal", "date": "2026-10-07", "version": null, "stance": "recommend", "quote": "FastFlowLM is what you're looking for", "excerpt": "FastFlowLM is what you're looking for - Lemonade does package it, as sibling comment says. https://fastflowlm.com/", "context": "", "use_case": "NPU model inference", "switched_from": "", "story": "Strands Decider 2B: a small, open-source, decision model"}, {"id": 776515794, "item_id": 48532237, "url": "https://news.ycombinator.com/item?id=48532237", "by": "jinnko", "date": "2026-06-14", "version": null, "stance": "recommend", "quote": "Check out FastFlowLM for AMD NPU support.", "excerpt": "Check out FastFlowLM for AMD NPU support.", "context": "", "use_case": "AMD NPU inference", "switched_from": "", "story": "Linux 7.1"}, {"id": 761845248, "item_id": 47615328, "url": "https://news.ycombinator.com/item?id=47615328", "by": "cpburns2009", "date": "2026-04-02", "version": null, "stance": "recommend", "quote": "The NPU works on Linux (Arch at least) on Strix Halo using FastFlowLM", "excerpt": "The NPU works on Linux (Arch at least) on Strix Halo using FastFlowLM [1]. Their NPU kernels are proprietary though (free up to a reasonable amount of commercial revenue). It's neat you can run some models basically for free (using NPU instead of CPU/GPU), but the performance is underwhelming. The target for NPUs is really low power devices, and not useful if you have an APU/GPU like Strix Halo. [1]: https://github.c", "context": "", "use_case": "", "switched_from": "", "story": "Lemonade by AMD: a fast and open source local LLM server using GPU and NPU"}, {"id": 761824353, "item_id": 47614022, "url": "https://news.ycombinator.com/item?id=47614022", "by": "dennemark", "date": "2026-04-02", "version": null, "stance": "neutral", "quote": "FLM is free for non-commercial use, however note that commercial licensing terms apply.", "excerpt": "Maybe you have seen NPU support via FLM already: https://lemonade-server.ai/flm_npu_linux.html \"FastFlowLM (FLM) support in Lemonade is in Early Access. FLM is free for non-commercial use, however note that commercial licensing terms apply. \"", "context": "", "use_case": "", "switched_from": "", "story": "Lemonade by AMD: a fast and open source local LLM server using GPU and NPU"}, {"id": 761822112, "item_id": 47613882, "url": "https://news.ycombinator.com/item?id=47613882", "by": "zozbot234", "date": "2026-04-02", "version": null, "stance": "warn", "quote": "NPU-accelerated kernels are proprietary binaries", "excerpt": "That won't give you NPU support, which relies on https://github.com/FastFlowLM/FastFlowLM . And that says \"NPU-accelerated kernels are proprietary binaries\", not open source.", "context": "", "use_case": "", "switched_from": "", "story": "Lemonade by AMD: a fast and open source local LLM server using GPU and NPU"}, {"id": 756219840, "item_id": 47263740, "url": "https://news.ycombinator.com/item?id=47263740", "by": "jakogut", "date": "2026-03-05", "version": null, "stance": "recommend", "quote": "a transformer framework called FastFlowLM [0] supporting XDNA 2 NPUs officially started supporting Linux", "excerpt": "You in fact can now! In the past week, a transformer framework called FastFlowLM [0] supporting XDNA 2 NPUs officially started supporting Linux. I posted it here the same day I found and started using it, to almost no reaction. [0] https://github.com/FastFlowLM https://fastflowlm.com/ https://huggingface.co/FastFlowLM", "context": "", "use_case": "running models on AMD NPUs", "switched_from": "", "story": "AMD will bring its \"Ryzen AI\" processors to standard desktop PCs for first time"}, {"id": 755653664, "item_id": 47228354, "url": "https://news.ycombinator.com/item?id=47228354", "by": "pstuart", "date": "2026-03-02", "version": null, "stance": "recommend", "quote": "Finally there's good news for Linux users", "excerpt": "Finally there's good news for Linux users: https://github.com/FastFlowLM/FastFlowLM", "context": "", "use_case": "running LLMs on AMD NPU", "switched_from": "", "story": "Welcome (back) to Macintosh"}, {"id": 754748992, "item_id": 47171812, "url": "https://news.ycombinator.com/item?id=47171812", "by": "jakogut", "date": "2026-02-26", "version": null, "stance": "neutral", "quote": "There's also a guide on the Framework community forums for getting this built and running on Arch Linux.", "excerpt": "There's also a guide on the Framework community forums for getting this built and running on Arch Linux. https://community.frame.work/t/guide-use-npu-xdna2-with-arch-linux-and-fastflowlm/80879", "context": "", "use_case": "", "switched_from": "", "story": "FastFlowLM (FLM) \u2013 Unlock Ryzen AI NPUs"}, {"id": 744841040, "item_id": 46552565, "url": "https://news.ycombinator.com/item?id=46552565", "by": "imtringued", "date": "2026-01-09", "version": null, "stance": "neutral", "quote": "https://fastflowlm.com/benchmarks/", "excerpt": "https://fastflowlm.com/benchmarks/ https://fastflowlm.com/assets/bench/gemma3-4b.png", "context": "", "use_case": "", "switched_from": "", "story": "Dell's CES 2026 chat was the most pleasingly un-AI briefing I've had in 5 years"}]}