# Fireworks AI 67 Hacker News commenters recommended Fireworks AI and 19 criticized it, November 2025 to November 2026. It was recommended in 11 of 13 months. It meets the Hacker Approved badge rule. **Verdict:** Recommended for fast, reliable hosting of open-weight models with strict data retention and US compliance, though quality and billing draw criticism. ## Why people recommend it - Strict data retention and privacy policies (7 comments): "Fireworks is one of the companies serving open models with solid performance and compliance/zero data retention sorted out." ([source](https://news.ycombinator.com/item?id=48446883)) "They are incredibly price competitive, have strict data retention policies, and in general are quite fast." ([source](https://news.ycombinator.com/item?id=48338579)) "together.ai, fireworks.ai have tons of deals that focus very much on data confidentiality." ([source](https://news.ycombinator.com/item?id=49622578)) - Wide range of open models (6 comments): "offer a wide range of models like gpt oss that are very capable and far cheaper than the models from big 3." ([source](https://news.ycombinator.com/item?id=48435889)) "You can pay for GLM 5.2 inference right now via Fireworks AI and presumably several dozen other providers." ([source](https://news.ycombinator.com/item?id=48615941)) "You can run them on the Together AIs or Firework AIs of this world." ([source](https://news.ycombinator.com/item?id=49410631)) - Competitive pricing for open models (4 comments): "offer a wide range of models like gpt oss that are very capable and far cheaper than the models from big 3." ([source](https://news.ycombinator.com/item?id=48435889)) "They are incredibly price competitive, have strict data retention policies, and in general are quite fast." ([source](https://news.ycombinator.com/item?id=48338579)) "Fireworks supports this model serverless for $1.20 per million tokens." ([source](https://news.ycombinator.com/item?id=46121078)) - Reliable with few rate limit issues (3 comments): "I've used GLM-5.2 a lot on fireworks and had never ever issues on rate limits." ([source](https://news.ycombinator.com/item?id=49071997)) "Use the model through a fast and reliable provider such as Fireworks directly" ([source](https://news.ycombinator.com/item?id=49899535)) "Yep, I have it running multiple OpenClaw instances all day. No issues." ([source](https://news.ycombinator.com/item?id=47821270)) ## What people warn about - Quality varies by model (3 comments): "Fireworks is a great provider but we still ran into issues here and there" ([source](https://news.ycombinator.com/item?id=49113390)) "Fireworks can be great, but it depends on the model and on the day." ([source](https://news.ycombinator.com/item?id=49660828)) "Fireworks has high variance amongst models and while some are served correctly; many of them are junk / broken and degraded" ([source](https://news.ycombinator.com/item?id=49658827)) - Limited model selection or LoRA support (3 comments): "My only gripe is their serverless offering for reranking has only 1 model which is Qwen 8b and it's quite expensive." ([source](https://news.ycombinator.com/item?id=49656891)) "Even the providers like fireworks which claim to let you deploy LORAs in a scalable way can't do it." ([source](https://news.ycombinator.com/item?id=47799096)) "the only models supporting LORA are like year old dense models that perform horribly on most tasks" ([source](https://news.ycombinator.com/item?id=47811584)) - Confusing or changing billing and subscriptions (3 comments): "subscription-based which is nicer for coding than e.g. Fireworks which is per-token billing" ([source](https://news.ycombinator.com/item?id=48741415)) "screwed up the onboarding, charged me money (when the account was supposed to be gratis)" ([source](https://news.ycombinator.com/item?id=50137415)) "I've been rugpulled multiple times on their Fire Pass subscription" ([source](https://news.ycombinator.com/item?id=49393619)) - Cache and token costs hard to compare (2 comments): "these zdr providers dont transparently disclose caching/tokens etc and so they end up being far more expensive than directly using DS" ([source](https://news.ycombinator.com/item?id=49124695)) "the cache rates for GLM 5.3 flash are quite a bit more expensive than for DeepSeek" ([source](https://news.ycombinator.com/item?id=49833278)) ## Alternatives hackers recommend - [Novita AI](https://hackerapproved.com/p/novita.md): 13 recommend - [Cloudflare Workers AI](https://hackerapproved.com/p/cloudflare-workers-ai.md): 7 recommend - [Chutes AI](https://hackerapproved.com/p/chutes.md): 6 recommend - [Telnyx Inference API](https://hackerapproved.com/p/telnyx-inference-api.md): 5 recommend ## Where to get it - [Visit the official site](https://fireworks.ai/) ## Cite as Hacker Approved, "Fireworks AI", https://hackerapproved.com/p/fireworks/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).