{"slug": "bonsai-2", "name": "Bonsai 2", "category": "ai", "category_name": "AI models and assistants", "description": "", "recommenders": 4, "commenters": 4, "warners": 0, "score": 3.948, "rising": 4.0, "first": "2026-09-18", "latest": "2026-10-31", "why": "For anyone who wants to try Bonsai 2 without setting up runtimes", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 4, "neutral": 0, "compare": 0, "warn": 0}, "spark": [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 2], "listed": true, "loved": [], "caveat": "", "short_quote": "For anyone who wants to try Bonsai 2 without setting up runtimes", "kind": "", "icon": false, "domain": "triangllabs.ai", "github_repo": null, "official_url": "https://triangllabs.ai", "pricing": "", "aliases": ["Bonsai 2"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 805142721, "item_id": 50321420, "url": "https://news.ycombinator.com/item?id=50321420", "by": "roenxi", "date": "2026-10-31", "stance": "recommend", "quote": "with the Bonsai 2 quantisations it is quite achievable", "excerpt": "That's pretty interesting, because with the Bonsai 2 quantisations it is quite achievable to get Qwen 27B running at 50 t/s on reasonably priced gaming machine [0]. Which in turn suggests that Fable 5 coding performance can be achieved locally with a bit of clever harnessing. Although the same harnessing should also improve Fable's performance. The important thing here is the cap for local performance looks like it c", "context": "", "use_case": "", "switched_from": "", "story": "Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hard"}, {"id": 804309681, "item_id": 50269355, "url": "https://news.ycombinator.com/item?id=50269355", "by": "vektormemory", "date": "2026-10-28", "stance": "recommend", "quote": "Or try these with your trained embedder", "excerpt": "Great idea, I do a lot of work testing small LLM's in this space and agree if you add some guardrails and train a small embedder, you won't need these large models, as your info is static, you can ship it all as one. Or try these with your trained embedder: https://ollama.com/library/minicpm-v4.6 https://ollama.com/library/bonsai-2", "context": "tests small LLMs a lot", "use_case": "", "switched_from": "", "story": "Survival LLM: AI assistant and emergency library that runs on your own computer"}, {"id": 797331216, "item_id": 49833201, "url": "https://news.ycombinator.com/item?id=49833201", "by": "ranger_danger", "date": "2026-09-24", "stance": "recommend", "quote": "keep in mind we are talking about Bonsai 2 and not the original Bonsai.", "excerpt": "Please cite a reputable source for this claim, and keep in mind we are talking about Bonsai 2 and not the original Bonsai. Even on the original one, the authors have replied to such issues with working solutions: https://github.com/PrismML-Eng/Bonsai-demo/issues/131", "context": "", "use_case": "", "switched_from": "", "story": "Best LLM for every budget, updated daily"}, {"id": 796066336, "item_id": 49754146, "url": "https://news.ycombinator.com/item?id=49754146", "by": "petrenk0n", "date": "2026-09-18", "stance": "recommend", "quote": "For anyone who wants to try Bonsai 2 without setting up runtimes", "excerpt": "For anyone who wants to try Bonsai 2 without setting up runtimes, downloading the right quant etc, try here - https://triangllabs.ai/otis", "context": "", "use_case": "running model without local setup", "switched_from": "", "story": "Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint"}]}], "caveats": [], "who": [{"by": "vektormemory", "context": "tests small LLMs a lot", "url": "https://news.ycombinator.com/item?id=50269355", "date": "2026-10-28"}], "tags": [], "use_cases": ["running model without local setup"], "compared": [], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=50320995", "title": "Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hard", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=50264853", "title": "Survival LLM: AI assistant and emergency library that runs on your own computer", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49830866", "title": "Best LLM for every budget, updated daily", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49746618", "title": "Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {}, "2026-02": {}, "2026-03": {}, "2026-04": {}, "2026-05": {}, "2026-06": {}, "2026-07": {}, "2026-08": {}, "2026-09": {"recommend": 2}, "2026-10": {"recommend": 2}}, "quotes": [{"id": 805142721, "item_id": 50321420, "url": "https://news.ycombinator.com/item?id=50321420", "by": "roenxi", "date": "2026-10-31", "stance": "recommend", "quote": "with the Bonsai 2 quantisations it is quite achievable", "excerpt": "That's pretty interesting, because with the Bonsai 2 quantisations it is quite achievable to get Qwen 27B running at 50 t/s on reasonably priced gaming machine [0]. Which in turn suggests that Fable 5 coding performance can be achieved locally with a bit of clever harnessing. Although the same harnessing should also improve Fable's performance. The important thing here is the cap for local performance looks like it c", "context": "", "use_case": "", "switched_from": "", "story": "Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hard"}, {"id": 804309681, "item_id": 50269355, "url": "https://news.ycombinator.com/item?id=50269355", "by": "vektormemory", "date": "2026-10-28", "stance": "recommend", "quote": "Or try these with your trained embedder", "excerpt": "Great idea, I do a lot of work testing small LLM's in this space and agree if you add some guardrails and train a small embedder, you won't need these large models, as your info is static, you can ship it all as one. Or try these with your trained embedder: https://ollama.com/library/minicpm-v4.6 https://ollama.com/library/bonsai-2", "context": "tests small LLMs a lot", "use_case": "", "switched_from": "", "story": "Survival LLM: AI assistant and emergency library that runs on your own computer"}, {"id": 797331216, "item_id": 49833201, "url": "https://news.ycombinator.com/item?id=49833201", "by": "ranger_danger", "date": "2026-09-24", "stance": "recommend", "quote": "keep in mind we are talking about Bonsai 2 and not the original Bonsai.", "excerpt": "Please cite a reputable source for this claim, and keep in mind we are talking about Bonsai 2 and not the original Bonsai. Even on the original one, the authors have replied to such issues with working solutions: https://github.com/PrismML-Eng/Bonsai-demo/issues/131", "context": "", "use_case": "", "switched_from": "", "story": "Best LLM for every budget, updated daily"}, {"id": 796066336, "item_id": 49754146, "url": "https://news.ycombinator.com/item?id=49754146", "by": "petrenk0n", "date": "2026-09-18", "stance": "recommend", "quote": "For anyone who wants to try Bonsai 2 without setting up runtimes", "excerpt": "For anyone who wants to try Bonsai 2 without setting up runtimes, downloading the right quant etc, try here - https://triangllabs.ai/otis", "context": "", "use_case": "running model without local setup", "switched_from": "", "story": "Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint"}]}