Nebius Token Factory (@nebiustf) / X

Nebius Token Factory

903 posts

Nebius Token Factory

@nebiustf

Own your intelligence. Precision-built inference for AI workloads at scale. Powered by @nebiusai | Discord: discord.com/invite/WJ2DUQR…

The Netherlands

tokenfactory.nebius.com/?utm_medium=X&…

Joined November 2024

Pinned
Nebius Token Factory
@nebiustf
Jun 25
Today we’re launching Custom Speculator Training in Nebius Token Factory. Train workload-specific draft models from your own data and deploy them alongside the base model. Generic drafters are a good start, but production traffic is rarely generic 👇
Nebius Token Factory
@nebiustf
Nov 5, 2025
⚡️Today marks a big milestone for Nebius. We’re launching Nebius Token Factory, the evolution of Nebius AI Studio, built to make open-source AI production-grade. Token Factory transforms raw open models into governed, scalable systems with dedicated inference, sub-second
00:00
Nebius Token Factory
@nebiustf
Aug 4, 2025
Fastest NVIDIA-GPU speeds on public boards: Llama 70B 245 tok/s, Qwen3-32B 212 tok/s (by @ArtificialAnlys) And we’ll make your fine-tune just as fast, by training a custom speculator for your model. Keep CUDA, bring LoRAs, ship. Zero custom-silicon lock-in.
Nebius Token Factory
@nebiustf
Oct 28, 2025
👀 Meet @NVIDIAAI Nemotron Nano 2 VL, now hosted on Nebius AI Studio - 10× higher throughput - Document + video intelligence - Open weights, open data - Ready for production Build multimodal assistants today → studio.nebius.com/playground?mod…
Nebius Token Factory
@nebiustf
Jan 29, 2025
You asked for it, we delivered: DeepSeek R1 is live on Nebius AI Studio. The reasoning power you want, the privacy you need, at prices you’ll love. Plus: DeepSeek V3, QwQ-32B, and Phi-4 join as well 🤝 Try it: 👇
Nebius Token Factory
@nebiustf
Feb 14, 2025
DeepSeek R1 just got faster 👀 Introducing our new high-performance endpoint: - Up to 60+ tokens/second - Advanced reasoning - Starting at $2/$6 per 1M tokens Try it now on Nebius AI Studio✨
Nebius Token Factory
@nebiustf
Aug 6, 2025
🚨 Now live on Nebius AI Studio: GPT-OSS-120B and GPT-OSS-20B: @OpenAI's new open-weight models. ✅ Instant access, no GPU setup ✅Playground + API ✅ Open source, zero-retention ✅Enterprise-grade, scalable inference Try them now 👇
Nebius Token Factory
@nebiustf
Aug 1, 2025
🚨 Now live: Qwen3-235B-A22B-2507 - 235B total params / 22B activated (MoE) - Structured, non-thinking mode: faster and more direct - Native 262K context window - Strong benchmark scores in math, logic, and coding - Zero-retention hosting A top-tier generalist LLM 👇
Nebius Token Factory
@nebiustf
Mar 4, 2025
Nebius AI Studio is now available on @huggingface 🤗 It means building with AI just got a whole lot easier. Access our powerful open-source models directly through the HF interface. No infrastructure setup, no deployment headaches. Just AI that works. (1/4)
Nebius Token Factory
@nebiustf
Aug 8, 2025
Speed isn’t luck, it’s good engineering. Nebius AI Studio is the fastest place to run DeepSeek R1 0528 (per @ArtificialAnlys) Blackwell B200 + speculative decoding + smart FP4 on our vertically integrated stack. Sub-second first token. Real throughput. 👇
Nebius Token Factory
@nebiustf
Jul 11, 2025
🤖 New on the Nebius blog: Agent 101 – Launching production‑grade AI agents at scale From proof‑of‑concept to production-level systems, we break down the full stack: ✅ LLMs ✅Frameworks: @crewAIInc, @langchain, @Google ADK, @AgnoAgi ✅Observability: @helicone_ai,
Nebius Token Factory
@nebiustf
May 1, 2025
Now serving: Qwen3 on Nebius AI Studio Inference endpoints for: - Qwen3-14B-base - Qwen3-235B-A22B-base - Qwen3-30B-A3B (base + fast) - Qwen3-32B-base 🛠️ Plus: fine-tuning now live for 10+ variants from 0.6B → 32B 💡 Build specialized copilots, agents & coders — deploy in
Nebius Token Factory
@nebiustf
Jul 1, 2025
Q2 at Nebius AI Studio: Wrapped! This quarter, we delivered more than 40 major updates to help AI builders and enterprises seamlessly build, customize and scale their AI solutions. 🌀 Adaptive elastic inference: Handle massive workloads effortlessly with our new adaptive
Nebius Token Factory
@nebiustf
May 23, 2025
Our friends at @huggingface just dropped Tiny Agents. Tool-savvy AI side-kicks in only ~70 Python lines, powered by @nebiusaistudio 🎉 Demo below: - searches Brave for “HF inference providers” - opens the first result - spits back the full provider list totally hands-free.
00:00