{"slug": "webllm", "name": "WebLLM", "category": "ai", "category_name": "AI tools and apps", "description": "", "recommenders": 4, "commenters": 7, "warners": 1, "score": 3.035, "rising": 0.02, "first": "2026-01-29", "latest": "2026-09-03", "why": "WebLLM https://github.com/mlc-ai/web-llm is more capable and will already work using WebGPU.", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 4, "neutral": 1, "compare": 1, "warn": 1}, "spark": [0, 0, 0, 0, 2, 1, 0, 0, 0, 0, 1, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "WebLLM https://github.com/mlc-ai/web-llm is more capable and will already work using WebGPU.", "kind": "", "icon": true, "domain": "mlc.ai", "github_repo": null, "official_url": "https://webllm.mlc.ai", "pricing": "", "aliases": ["WebLLM"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 792786816, "item_id": 49549176, "url": "https://news.ycombinator.com/item?id=49549176", "by": "karuppiah7890", "date": "2026-09-03", "version": null, "stance": "recommend", "quote": "I love it! It works fine for me", "excerpt": "I love it! It works fine for me on Google Chrome on my MacBook Pro 14-inch M5 16GB RAM and 10 Cores CPU and 10 Cores GPU, with Metal 4 support", "context": "", "use_case": "in-browser LLM inference", "switched_from": "", "story": "WebLLM: high-performance in-browser LLM inference engine"}, {"id": 767400721, "item_id": 47962545, "url": "https://news.ycombinator.com/item?id=47962545", "by": "ilaksh", "date": "2026-04-30", "version": null, "stance": "recommend", "quote": "WebLLM https://github.com/mlc-ai/web-llm is more capable and will already work using WebGPU.", "excerpt": "The Prompt API has some advantages like being a little simpler for some things and some potential to standardize a little bit more in some way, but it looks like from this that it will be delayed unfortunately. However, WebLLM https://github.com/mlc-ai/web-llm is more capable and will already work using WebGPU.", "context": "", "use_case": "", "switched_from": "", "story": "Mozilla's Opposition to Chrome's Prompt API"}, {"id": 756336480, "item_id": 47271030, "url": "https://news.ycombinator.com/item?id=47271030", "by": "darkvertex", "date": "2026-03-05", "version": null, "stance": "recommend", "quote": "Or why not stay fully local with WebLLM", "excerpt": "Or why not stay fully local with WebLLM... https://webllm.mlc.ai", "context": "", "use_case": "Local LLM in browser", "switched_from": "", "story": "Show HN: PageAgent, A GUI agent that lives inside your web app"}]}], "caveats": [{"label": "", "summary": "", "quotes": [{"id": 792600752, "item_id": 49537547, "url": "https://news.ycombinator.com/item?id=49537547", "by": "TekMol", "date": "2026-09-02", "version": null, "stance": "warn", "quote": "WebGPUNotAvailableError", "excerpt": "This seems to be the demo: https://chat.webllm.ai/ I am getting: WebGPUNotAvailableError: WebGPU is not supported in your current environment, but it is necessary to run the WebLLM engine. On both, FireFox and Chromium on Linux.", "context": "", "use_case": "", "switched_from": "", "story": "WebLLM: high-performance in-browser LLM inference engine"}]}], "who": [], "tags": [], "use_cases": ["in-browser llm inference", "local llm in browser"], "compared": [{"slug": "onnx-runtime", "name": "ONNX Runtime", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=49536411", "title": "WebLLM: high-performance in-browser LLM inference engine", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=47264138", "title": "Show HN: PageAgent, A GUI agent that lives inside your web app", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=47959463", "title": "Mozilla's Opposition to Chrome's Prompt API", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=47233205", "title": "Show HN: I built an AI data analyst that never sees your data", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46812583", "title": "Show HN: Kandle \u2013 A WebGPU-based ML library written from scratch in JavaScript", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {"compare": 1}, "2026-02": {}, "2026-03": {"neutral": 1, "recommend": 2}, "2026-04": {"recommend": 1}, "2026-05": {}, "2026-06": {}, "2026-07": {}, "2026-08": {}, "2026-09": {"warn": 1, "recommend": 1}, "2026-10": {}}, "versions": [], "version_unnamed": 0, "family_ref": null, "runs_on": null, "lines": [], "quotes": [{"id": 792786816, "item_id": 49549176, "url": "https://news.ycombinator.com/item?id=49549176", "by": "karuppiah7890", "date": "2026-09-03", "version": null, "stance": "recommend", "quote": "I love it! It works fine for me", "excerpt": "I love it! It works fine for me on Google Chrome on my MacBook Pro 14-inch M5 16GB RAM and 10 Cores CPU and 10 Cores GPU, with Metal 4 support", "context": "", "use_case": "in-browser LLM inference", "switched_from": "", "story": "WebLLM: high-performance in-browser LLM inference engine"}, {"id": 792600752, "item_id": 49537547, "url": "https://news.ycombinator.com/item?id=49537547", "by": "TekMol", "date": "2026-09-02", "version": null, "stance": "warn", "quote": "WebGPUNotAvailableError", "excerpt": "This seems to be the demo: https://chat.webllm.ai/ I am getting: WebGPUNotAvailableError: WebGPU is not supported in your current environment, but it is necessary to run the WebLLM engine. On both, FireFox and Chromium on Linux.", "context": "", "use_case": "", "switched_from": "", "story": "WebLLM: high-performance in-browser LLM inference engine"}, {"id": 767400721, "item_id": 47962545, "url": "https://news.ycombinator.com/item?id=47962545", "by": "ilaksh", "date": "2026-04-30", "version": null, "stance": "recommend", "quote": "WebLLM https://github.com/mlc-ai/web-llm is more capable and will already work using WebGPU.", "excerpt": "The Prompt API has some advantages like being a little simpler for some things and some potential to standardize a little bit more in some way, but it looks like from this that it will be delayed unfortunately. However, WebLLM https://github.com/mlc-ai/web-llm is more capable and will already work using WebGPU.", "context": "", "use_case": "", "switched_from": "", "story": "Mozilla's Opposition to Chrome's Prompt API"}, {"id": 756341248, "item_id": 47271328, "url": "https://news.ycombinator.com/item?id=47271328", "by": "simon_luv_pho", "date": "2026-03-05", "version": null, "stance": "recommend", "quote": "", "excerpt": "That looks great! I also thought about calling the Gemini nano model embedded into Chrome (only extensions can do that). But after some testing on smaller models I found that anything smaller than 9b can\u2019t really handle the complex tool call schema I use. Qwen3.5 4b is quite good but still gives messy json quite often. But it\u2019s very promising! Maybe after one more model iteration or some fine-toning we can go fully e", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: PageAgent, A GUI agent that lives inside your web app"}, {"id": 756336480, "item_id": 47271030, "url": "https://news.ycombinator.com/item?id=47271030", "by": "darkvertex", "date": "2026-03-05", "version": null, "stance": "recommend", "quote": "Or why not stay fully local with WebLLM", "excerpt": "Or why not stay fully local with WebLLM... https://webllm.mlc.ai", "context": "", "use_case": "Local LLM in browser", "switched_from": "", "story": "Show HN: PageAgent, A GUI agent that lives inside your web app"}, {"id": 755731636, "item_id": 47233227, "url": "https://news.ycombinator.com/item?id=47233227", "by": "david-rodriguez", "date": "2026-03-03", "version": null, "stance": "neutral", "quote": "For the free tier, even the AI runs locally (WebLLM/Ollama)", "excerpt": "Hey HN, I'm the builder. The core insight behind this is simple: AI doesn't need your data to analyze it \u2014 it only needs your schema. When you ask \"what's the average order value by region?\", the LLM just needs to know there's an order_value column (decimal) and a region column (varchar). It generates SQL from that. DuckDB-WASM executes the query locally in your browser. The AI never sees a single row. Architecture: ", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: I built an AI data analyst that never sees your data"}, {"id": 749001410, "item_id": 46812588, "url": "https://news.ycombinator.com/item?id=46812588", "by": "finalkk", "date": "2026-01-29", "version": null, "stance": "compare", "quote": "They are incredible for production", "excerpt": "Hi HN, I\u2019ve spent the last 3 months building Kandle because I was frustrated with the status quo of ML on the Web. Currently, if you want to run models in the browser, you\u2019re mostly stuck with ONNX Runtime or WebLLM. They are incredible for production, but they are \"Blackboxes.\" You have almost zero control over the intermediate tensors, and implementing custom logic or hooks between layers is a nightmare. I missed t", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: Kandle \u2013 A WebGPU-based ML library written from scratch in JavaScript"}]}