Tom Humphrey
527 posts

Tom Humphrey
@TomHums
investing with @blackbirdvc. always composting. love a good dad joke

Introducing Unsloth Studio ✨ A new open-source web UI to train and run LLMs. • Run models locally on Mac, Windows, Linux • Train 500+ models 2x faster with 70% less VRAM • Supports GGUF, vision, audio, embedding models • Auto-create datasets from PDF, CSV, DOCX • Self-healing tool calling and code execution • Compare models side by side + export to GGUF GitHub: github.com/unslothai/unsl… Blog and Guide: unsloth.ai/docs/new/studio Available now on Hugging Face, NVIDIA, Docker and Colab.





A humanoid robot played a soccer match. 11 kick attempts. 8 falls. 1 overtime goal. 9% shot conversion. We didn't watch the footage. Just uploaded the data and asked. TryAlloy for free today. Link below!

Super proud to unveil our AI marketing platform, Checkmate Performance Network. We've been working with 250+ brands (Everlane, Stanley +) to drive sales. That momentum helped Checkmate cross a $10M run rate in <1 year, up 900%+ in < 12 months & we’re profitable! Stoked for 2026








Leonardo AI, part of Canva, has released Lucid Origin, debuting at #6 in the Artificial Analysis Image Leaderboard, delivering quality comparable to FLUX.1 Kontext and marking a major upgrade over its Phoenix models Leonardo AI, the Australian generative media company acquired by Canva in 2024, has launched Lucid Origin in both Ultra and Fast variants. Lucid Origin Ultra delivers quality on par with Black Forest Labs’ FLUX.1 Kontext [max], while the Fast variant rivals FLUX.1 Kontext [pro]. Unlike competitors that price per image, Leonardo uses a subscription-based API credit system with monthly plans and optional top-ups. On the highest value $299/month tier, Lucid Origin Ultra costs ~$87 per 1k images, positioning it alongside other premium models such as Seedream 3.0 ($30/1k), Imagen 4 ($40/1k), FLUX.1 Kontext [max] ($80/1k), and GPT-4o (~$167/1k). The Fast variant at ~$18/1k, offers a more cost-efficient alternative. Lucid Origin Ultra generates images in ~22s, with the Fast variant completing in ~12s. By comparison, Seedream 3.0 and FLUX.1 Kontext [max] generate in ~5s, Imagen 4 Ultra in ~12s, and GPT-4o in ~51s. Leonardo AI’s models are available through its generative media studio app and developer API, alongside its recently launched Motion 2.0 video models. See below for comparisons between Lucid Origin and other leading models in our Artificial Analysis Image Arena 🧵





Spice AI Partners With @databricks to Extend Operational Data & AI Capabilities to Real-Time Applications and Agents Spice AI’s new integration with Databricks enables customers to build high-performance, secure, and business-ready AI applications and agents for the enterprise. Stop by and visit our booth this week at the #DataAISummit (D114). Read the full press release: bit.ly/3Zqr9AL









