{"slug": "workers-ai", "name": "Workers AI", "category": "ai", "category_name": "AI tools and apps", "description": "", "recommenders": 4, "commenters": 11, "warners": 2, "score": 3.249, "rising": 0.02, "first": "2025-11-17", "latest": "2026-10-20", "why": "Replicate is basically just a Worker AI endpoint and easy to maintain", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 4, "neutral": 5, "compare": 0, "warn": 2}, "spark": [1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 1], "listed": true, "loved": [], "caveat": "", "short_quote": "Replicate is basically just a Worker AI endpoint and easy to maintain", "kind": "", "icon": true, "domain": "cloudflare.com", "github_repo": null, "official_url": "https://developers.cloudflare.com/workers-ai/", "pricing": "", "aliases": ["Cloudflare Workers AI", "Workers AI"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 802649349, "item_id": 50165584, "url": "https://news.ycombinator.com/item?id=50165584", "by": "horsawlarway", "date": "2026-10-20", "version": null, "stance": "recommend", "quote": "if cloudflare is hosting it, we're exploring it", "excerpt": "Nobody meets expectations of a 2 trillion dollar IPO selling a fungible commodity like tokens for old models. And that's the rub - it doesn't matter if inference is profitable, the US labs have promised the whole kitchen sink, and taken serious investment on that pitch. They won't make it back in a cut-throat market for inference. They have to continue pitching the idea that only their models are worth using, for qua", "context": "", "use_case": "", "switched_from": "", "story": "DeepInfer: High-performance inference framework for DeepSeek models"}, {"id": 748397248, "item_id": 46774828, "url": "https://news.ycombinator.com/item?id=46774828", "by": "TurdF3rguson", "date": "2026-01-26", "version": null, "stance": "recommend", "quote": "without exceeding daily free tier.", "excerpt": "Those text embeddings are dirt cheap. You can do around 1M titles on the cloudflare embedding model I used last time without exceeding daily free tier.", "context": "", "use_case": "text embeddings", "switched_from": "", "story": "Find 'Abbey Road when type 'Beatles abbey rd': Fuzzy/Semantic search in Postgres"}, {"id": 735343202, "item_id": 45958950, "url": "https://news.ycombinator.com/item?id=45958950", "by": "NicoJuicy", "date": "2025-11-17", "version": null, "stance": "recommend", "quote": "Replicate is basically just a Worker AI endpoint and easy to maintain", "excerpt": "I don't think that much will change regarding to Replicate's API because of Cloudflare. It's specifically mentioned in Replicate's blog post and it's also not in their best interest. What's important to know ( I think). Recently, Cloudflare released a blog post of Omni for AI inference. I think they performance tuned it better than other providers. So their costs per inference drops down a lot ( https://blog.cloudfla", "context": "", "use_case": "", "switched_from": "", "story": "Replicate is joining Cloudflare"}]}], "caveats": [{"label": "", "summary": "", "quotes": [{"id": 763838804, "item_id": 47739925, "url": "https://news.ycombinator.com/item?id=47739925", "by": "mrnoname", "date": "2026-04-12", "version": null, "stance": "warn", "quote": "Workers AI has an OpenAI-compatible API so I expected it to just work with Codex. Nope.", "excerpt": "These days I reach for Codex CLI more than Claude Code. Not just for token economics, Codex has been catching up fast and is already ahead in some areas. It was open source from day one, arguably better code quality than the leaked Claude Code source, yet nobody cared. It's been improving quietly under the radar. I subscribe to ChatGPT Pro, but Ralph + Fast Mode can still burn through tokens. My company has a sizable", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: Codex Workers AI Proxy \u2013 Use Cloudflare Workers AI models in Codex CLI"}, {"id": 735284433, "item_id": 45955277, "url": "https://news.ycombinator.com/item?id=45955277", "by": "pzo", "date": "2025-11-17", "version": null, "stance": "warn", "quote": "But I think it didn't catch up and they have there very old models.", "excerpt": "They already did and they had Worker AI for quite a while [0]. But I think it didn't catch up and they have there very old models. Overall I think it's a good acquisition for both of them. Replicate would have to have more competitive pricing to compete with fal.ai [0] https://developers.cloudflare.com/workers-ai/models/", "context": "", "use_case": "", "switched_from": "", "story": "Replicate is joining Cloudflare"}]}], "who": [], "tags": [], "use_cases": ["serverless ai inference", "open-weights model hosting", "hosted ai model inference", "text embeddings"], "compared": [{"slug": "kimi", "name": "Kimi", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "openai-agents-sdk", "name": "OpenAI Agents SDK", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "replicate", "name": "Replicate", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [{"id": 767635568, "item_id": 47977223, "url": "https://news.ycombinator.com/item?id=47977223", "by": "kflansburg", "date": "2026-05-01", "version": null, "stance": "recommend", "quote": "Cloudflare is building across the entire AI stack.", "excerpt": "Cloudflare | Systems or ML Engineer, Workers AI | Austin, TX or London, UK or San Francisco, CA (Hybrid) | Full-Time | https://cloudflare.com Cloudflare is building across the entire AI stack. Here are some exciting things that we have launched recently: - Kimi k2.5: https://blog.cloudflare.com/workers-ai-large-models/ - Kimi Performance Improvements: https://blog.cloudflare.com/high-performance-llms/ - Kimi k2.6 Spe", "context": "", "use_case": "", "switched_from": "", "story": "Ask HN: Who is hiring? (May 2026)"}, {"id": 761662544, "item_id": 47603909, "url": "https://news.ycombinator.com/item?id=47603909", "by": "kflansburg", "date": "2026-04-01", "version": null, "stance": "recommend", "quote": "Senior Systems Engineer, Workers AI", "excerpt": "Cloudflare | Systems or ML Engineer, Workers AI | Austin, TX or London, UK or San Francisco, CA (Hybrid) | Full-Time | https://cloudflare.com Cloudflare is building across the entire AI stack. Here are some exciting things that we have launched recently: - Release Cog v0.17: https://github.com/replicate/cog/releases/tag/v0.17.0 - Dynamic Worker Sandboxes: https://blog.cloudflare.com/dynamic-workers/ - Kimi K2.5: http", "context": "", "use_case": "", "switched_from": "", "story": "Ask HN: Who is hiring? (April 2026)"}], "threads": [{"url": "https://news.ycombinator.com/item?id=47792538", "title": "Cloudflare's AI Platform: an inference layer designed for agents", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=45953702", "title": "Replicate is joining Cloudflare", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=50163277", "title": "DeepInfer: High-performance inference framework for DeepSeek models", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=50031614", "title": "Talorys \u2013 A self-hosted personal AI agent on Cloudflare's free tier", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49947690", "title": "Cube Rule food classifier using Cloudflare Clef classification model", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49158581", "title": "Smaller, faster, safer: running Kimi and GLM at scale", "mentions": 1}], "sentiment_by_month": {"2025-11": {"warn": 1, "recommend": 1}, "2025-12": {}, "2026-01": {"recommend": 1}, "2026-02": {}, "2026-03": {}, "2026-04": {"warn": 1, "neutral": 2}, "2026-05": {"recommend": 1}, "2026-06": {}, "2026-07": {}, "2026-08": {"neutral": 1}, "2026-09": {}, "2026-10": {"neutral": 2, "recommend": 1}}, "versions": [], "version_unnamed": 0, "quotes": [{"id": 802649349, "item_id": 50165584, "url": "https://news.ycombinator.com/item?id=50165584", "by": "horsawlarway", "date": "2026-10-20", "version": null, "stance": "recommend", "quote": "if cloudflare is hosting it, we're exploring it", "excerpt": "Nobody meets expectations of a 2 trillion dollar IPO selling a fungible commodity like tokens for old models. And that's the rub - it doesn't matter if inference is profitable, the US labs have promised the whole kitchen sink, and taken serious investment on that pitch. They won't make it back in a cut-throat market for inference. They have to continue pitching the idea that only their models are worth using, for qua", "context": "", "use_case": "", "switched_from": "", "story": "DeepInfer: High-performance inference framework for DeepSeek models"}, {"id": 800512032, "item_id": 50032002, "url": "https://news.ycombinator.com/item?id=50032002", "by": "martheen", "date": "2026-10-10", "version": null, "stance": "neutral", "quote": "10k neuron is roughly 0.11 USD for the usage", "excerpt": "10k neuron is roughly 0.11 USD for the usage https://developers.cloudflare.com/workers-ai/platform/pricing/", "context": "", "use_case": "serverless AI inference", "switched_from": "", "story": "Talorys \u2013 A self-hosted personal AI agent on Cloudflare's free tier"}, {"id": 799163058, "item_id": 49947691, "url": "https://news.ycombinator.com/item?id=49947691", "by": "murderszn", "date": "2026-10-03", "version": null, "stance": "neutral", "quote": "Clef Flash running via Cloudflare Workers AI", "excerpt": "I built a small demo that classifies food using the Cube Rule rather than asking a general model to write an essay on whether a hot dog is a sandwich. The Cube Rule categorizes food by the location of structural starch relative to the six faces of a cube: 0 Starchless (Salad): Steak, soup 1 Face (Toast): Pizza, nigiri 2 Faces (Sandwich): Burger, unfolded quesadilla 3 Faces (Taco): Hot dog, hard-shell taco 4 Faces (Su", "context": "", "use_case": "", "switched_from": "", "story": "Cube Rule food classifier using Cloudflare Clef classification model"}, {"id": 786660208, "item_id": 49166263, "url": "https://news.ycombinator.com/item?id=49166263", "by": "zx76", "date": "2026-08-04", "version": null, "stance": "neutral", "quote": "Cloudflare \u2022 Prompts are retained for unknown period", "excerpt": "Seems like the relevant page is: https://developers.cloudflare.com/workers-ai/platform/data-usage/ and their privacy policy also applies: https://www.cloudflare.com/privacypolicy/ \"Your inputs (e.g., text prompts, image submissions, audio files, etc.), outputs (e.g., generated text/images, translations, etc.), embeddings, and training data constitute Customer Content. For Workers AI: * You own, and are responsible fo", "context": "", "use_case": "", "switched_from": "", "story": "Smaller, faster, safer: running Kimi and GLM at scale"}, {"id": 768414864, "item_id": 48025929, "url": "https://news.ycombinator.com/item?id=48025929", "by": "andruby", "date": "2026-05-05", "version": null, "stance": "recommend", "quote": "", "excerpt": "They do indeed. See https://developers.cloudflare.com/workers-ai/models/ They seem to allow some free usage without user account. Do they list limits anywhere?", "context": "", "use_case": "Open-weights model hosting", "switched_from": "", "story": "Accelerating Gemma 4: faster inference with multi-token prediction drafters"}, {"id": 764761536, "item_id": 47797596, "url": "https://news.ycombinator.com/item?id=47797596", "by": "james2doyle", "date": "2026-04-16", "version": null, "stance": "neutral", "quote": "I find it really confusing that the worker AI models on here", "excerpt": "I find it really confusing that the worker AI models on here: https://developers.cloudflare.com/workers-ai/models/ do not have full overlap with the ones on here: https://developers.cloudflare.com/ai/models/ Yes, you can see the same \"hosted\" ones on there, but when you look at the models endpoint, there are much less options at the \"workers-ai/*\" namespace. Is that intentional?", "context": "", "use_case": "Hosted AI model inference", "switched_from": "", "story": "Cloudflare's AI Platform: an inference layer designed for agents"}, {"id": 764690272, "item_id": 47793142, "url": "https://news.ycombinator.com/item?id=47793142", "by": "yoavm", "date": "2026-04-16", "version": null, "stance": "neutral", "quote": "Workers AI pricing is this", "excerpt": "Workers AI pricing is this: https://developers.cloudflare.com/workers-ai/platform/pricing/", "context": "", "use_case": "", "switched_from": "", "story": "Cloudflare's AI Platform: an inference layer designed for agents"}, {"id": 763838804, "item_id": 47739925, "url": "https://news.ycombinator.com/item?id=47739925", "by": "mrnoname", "date": "2026-04-12", "version": null, "stance": "warn", "quote": "Workers AI has an OpenAI-compatible API so I expected it to just work with Codex. Nope.", "excerpt": "These days I reach for Codex CLI more than Claude Code. Not just for token economics, Codex has been catching up fast and is already ahead in some areas. It was open source from day one, arguably better code quality than the leaked Claude Code source, yet nobody cared. It's been improving quietly under the radar. I subscribe to ChatGPT Pro, but Ralph + Fast Mode can still burn through tokens. My company has a sizable", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: Codex Workers AI Proxy \u2013 Use Cloudflare Workers AI models in Codex CLI"}, {"id": 748397248, "item_id": 46774828, "url": "https://news.ycombinator.com/item?id=46774828", "by": "TurdF3rguson", "date": "2026-01-26", "version": null, "stance": "recommend", "quote": "without exceeding daily free tier.", "excerpt": "Those text embeddings are dirt cheap. You can do around 1M titles on the cloudflare embedding model I used last time without exceeding daily free tier.", "context": "", "use_case": "text embeddings", "switched_from": "", "story": "Find 'Abbey Road when type 'Beatles abbey rd': Fuzzy/Semantic search in Postgres"}, {"id": 735343202, "item_id": 45958950, "url": "https://news.ycombinator.com/item?id=45958950", "by": "NicoJuicy", "date": "2025-11-17", "version": null, "stance": "recommend", "quote": "Replicate is basically just a Worker AI endpoint and easy to maintain", "excerpt": "I don't think that much will change regarding to Replicate's API because of Cloudflare. It's specifically mentioned in Replicate's blog post and it's also not in their best interest. What's important to know ( I think). Recently, Cloudflare released a blog post of Omni for AI inference. I think they performance tuned it better than other providers. So their costs per inference drops down a lot ( https://blog.cloudfla", "context": "", "use_case": "", "switched_from": "", "story": "Replicate is joining Cloudflare"}, {"id": 735284433, "item_id": 45955277, "url": "https://news.ycombinator.com/item?id=45955277", "by": "pzo", "date": "2025-11-17", "version": null, "stance": "warn", "quote": "But I think it didn't catch up and they have there very old models.", "excerpt": "They already did and they had Worker AI for quite a while [0]. But I think it didn't catch up and they have there very old models. Overall I think it's a good acquisition for both of them. Replicate would have to have more competitive pricing to compete with fal.ai [0] https://developers.cloudflare.com/workers-ai/models/", "context": "", "use_case": "", "switched_from": "", "story": "Replicate is joining Cloudflare"}]}