Image created with Gemini. Image prompt: A horizontal 1920s Dada Merz collage on aged kraft board, layered with torn ledger pages listing model names and download numbers, small naive cut-paper smiling faces with open hands taped and pinned in overlapping rows like community trading cards, with the title ‘HuggingFace’ spelled in mismatched cut-out letterpress typefaces glued at slight angles across the top. Aged newsprint cream, faded vermilion, ink black, muted slate blue palette, flat even lighting, visible paper fiber and glue stains, photographed straight on as a physical artwork.
been testing GLM 5.2 directly inside Claude Code. it is a really good model here’s an ultra simple way to vibe check it via @huggingface “` export ANTHROPIC_BASE_URL=”https://t.co/q5zcSfYxpH” export ANTHROPIC_AUTH_TOKEN=”${HF_TOKEN}” claude –model “zai-org/GLM-5.2″ “`”
https://x.com/multimodalart/status/2068026613787217943
At @huggingface, we rely on Moon Bot religiously. It’s our async, Slack-based coding agent powered by Pi Agent+ HF Buckets. It’s connected to all of our services: > GithHub – Read Code / PRs / Issues > Athena – Query logs via AWS CLI > @PlausibleHQ – Web Traffic / Analytics >”
https://x.com/calebfahlgren/status/2069768499510013978
Here’s our new, tiniest model: LFM2.5-230M! 🥳 We went even smaller to power ultra-low latency use cases like e-commerce and robotics. Here’s a demo of LFM2.5-230M running on a Unitree G1, decomposing user prompts into a sequence of tool calls. Available today on @huggingface!”
https://x.com/maximelabonne/status/2070149175006617682
At Hugging Face we’ve been building our own agent that we use via Slack (Moon Bot). Honestly, building your own is quite simple and you’ll be happy you did: any model you want (self-hosted if needed), fully customizable to your stack (drop in a skill file and it can use any”
https://x.com/victormustar/status/2069696147526947290
Build real agentic apps using CUGA: two dozen working examples on a lightweight harness
https://huggingface.co/blog/ibm-research/cuga-apps
Mistral claims SOTA performance on OlmOCRBench, a popular optical character recognition benchmark, but that isn’t the case. We have a public leaderboard on @huggingface, where Mistral OCR 4 currently ranks #3, behind open models like Chandra OCR 2 by @datalabto”
https://x.com/NielsRogge/status/2069432947711652210
Accelerating Transformers Fine-Tuning with NVIDIA NeMo AutoModel
https://huggingface.co/blog/nvidia/accelerating-fine-tuning-nvidia-nemo-automodel
Welcome to Open Source AI: Run Your Own Models Locally”
https://x.com/huggingface/status/2070160187751850242
you can download them now from Huggingface: –
https://t.co/KcCsO5EiVq -“
https://x.com/krea_ai/status/2069435601078935601
The rise of MoE models introduced new challenges in training, and @huggingface’s Transformers v5 brought first-class support for solving them. Now, NeMo AutoModel builds on top of v5. Part of the NeMo framework for building models at scale, NeMo AutoModel brings optimizations to”
https://x.com/NVIDIAAI/status/2069813582825418828
Leave a Reply