Today we're announcing our $20 M Series A, co-led by @AMDVentures and @AirbusVentures, with @BMWiVentures, @Kickstartph, @PanacheVC, and @Wavemaker_VC. We're building the neutral layer for open models — so the future doesn't run on a few labs, a few chips, or a few clouds. THE FUTURE CAN’T BE BLOCKED. Any Model. Anywhere. Anyone. Yours.

BREAKING: Fable is dead. Press F(able) to pay respects. This has never happened to AI models hosted on Featherless, just saying.

Mythos doesn't even know what a token budget is it just looks at you like this.

Open WebUI on Featherless, it's that easy!

I read the Claude Fable blog post and immediately ran out of usage. That won't happen at Featherless, because right now you can sign up for our fixed priced unlimited token plans and easily access thousands of models.

Mythos sucks man what the hell is this.

Ask away your controversial questions via Featherless AI. We don't read your logs, or gatekeep any models.

Featherless 🤝 Respan We’re thrilled to announce that we’ll be joining Respan Gateway as a launch partner. Teams on Respan can now access Featherless AI for serverless open-weight inference, zero infra required. If you're building on Respan, we’ll be there for you. respan.ai/ai-gateway

Interested in learning more about the @NousResearch Hermes Agent? At Featherless we make it trivial to run and deploy agents safely and securely. Below is a quick demo.

We're working hard to acquire new GPUs for our customers... but this server room is confusing to navigate through. #backrooms

If it's on Hugging Face, you can easily run your gigaslop_27b_q8 model on Featherless via our API. Our mission is to make all LLMs globally accessible, even the edge cases.

We'd like to correct a mistake that we made. On our website, we stated that we serve over 30,000 open source models to the world. This was misleading -- we currently host a bit over open source 40,000 models. We'll do better next time.

Summer is here, and so is the 12th @AITinkerers Prague, June meetup with @posthog. We’re getting ready for an evening of demos, experiments, and conversations with Prague's builder community, and we want 🫵 you to show what you’ve been cooking (apply to RSVP and/or demo your project below)! Join us on June 12th for a dev meetup with tinkerers like yourself, founders, engineers, researchers - all in one room! The best part of the community has always been seeing what people are actually building, whether you’re deep into agentic pipelines, building AI products, experimenting with open-source tools, or simply curious about what others are working on, we'd love to have you there. Huge thanks to our partners @ValkaAIonX , @posthog , @FeatherlessAI , and @TensorVC for supporting the Prague AI Tinkerers community. See you on June 12th 👋

The truth behind closed source AI, and this is just what Sam's willing to say publicly. Open source will win.

We've had questions regarding SOC-2 and HIPAA certifications -- we're actively working with partners to explore both! (Rest assured, we're NOT using @getdelve).

You can now run our Nebula-S-V1 on @FeatherlessAI AGI starts here 🧠 featherless.ai/models/decomp…

🛑[LIVE] on day 2 of @aiDotEngineer 🇸🇬 @picocreator from @FeatherlessAI on wtf do people use open models for Open-source AI got so good it runs on your laptop now, which means the excuse economy is officially over. “Just start building”

Proud to announce that @FeatherlessAI is backing the WHYC demo night in Paris. 1 week to go. Trust me, this is going to be best event for founders Paris has ever witnessed. CC: @whycshow @0xratnakar @tushaarmehtaa

Introducing KeyValueMeans Our new AI architecture, for choosing between - Fixed Cost Recurrent State: ie. RWKV, GDN, Mamba - Quadratic Cost Scaling State: Attention - & a middle ground in between... What if Transformer attention, could expand on demand, instead of every token?