AI BriefingHuggingFace Blog news
The latest AI and developer news from HuggingFace Blog, with short summaries.
Visit website →TII Releases Falcon-ASR: 1.6B Parameter Model for Arabic and English Speech Recognition
The model achieves a 20.92% average Word Error Rate on Arabic benchmarks, outperforming the previous best published result by 2.25 percentage points.
NVIDIA Fine-Tunes Nemotron for Gold-Level Results at IOI and IMO 2026
Nemotron-3-Ultra-CC scored 535.4/600 at IOI 2026, surpassing the top human score, while a specialized IMO system achieved 30/42 points.
TII Releases Falcon-Emirati-7B: A Dialect-Specialized LLM for Emirati Arabic
The new 7B model achieves 84.83% on the Alyah benchmark and significantly outperforms larger general-purpose models in dialect fidelity and cultural understanding.
PickMicrosoft and Hugging Face Release ThinkingBox Benchmark Evaluating AI Agents by Backend State
The new benchmark evaluates 507 stateful business workflows by verifying final database states rather than tool call success, revealing that 67% of agent failures occur despite clean termination.
PickAi2 Open-Sources AstaBrief 8B for Fast, Cost-Effective Research Report Generation
The 8B model, based on Qwen3-8B, generates reports in an average of 51.1 seconds compared to 178.5 seconds for the previous Claude-based Thinking mode.
ServiceNow Releases AutoSynthData for Generating Enterprise Agent Training Data
The framework improves Pass@1 scores by up to 8.4 percentage points in ITSM and Hybrid enterprise environments by targeting model capability gaps.
PickAi2 Releases Olmo-core 3: Open Infrastructure Benchmarked for Trillion-Parameter MoE Training
Ai2 releases Olmo-core 3, an open-source training framework designed for large MoE models, achieving 2.7x throughput improvement in preliminary tests and benchmarked at over 1 trillion parameters.
Hugging Face Launches Open TTS Leaderboard for Scalable Multilingual and Voice Cloning Evaluation
The new leaderboard uses objective metrics like WER and RTFx to evaluate open-source TTS models, reducing assessment time from weeks to hours.
NVIDIA Releases Kumo Tabular: Open Foundation Model for Zero-Shot Tabular Prediction
Kumo Tabular achieves state-of-the-art accuracy on TabArena with ELO 1950 while being 17x faster than LimiX-2, trained entirely on synthetic data.
Multiverse Computing Introduces ProvenanceGuard for Source-Aware Verification of MCP Agents
The new verification layer caught 138 of 139 expert-flagged unsupported claims in medical agent tests while identifying the correct source for 86% of claims with identifiable sources.
H Company releases Holo4 generalist computer-use agent models
The new Holo4 models, available in 27B dense and 35B-A3B MoE sizes, interact with software across GUI, code, MCP, and API interfaces.
Liquid AI Releases LFM2.5-VL-DSpark: An Experimental Draft Model for Faster Vision-Language Inference
The new 280M-parameter draft model accelerates LFM2.5-VL-3B inference by up to 3.13x on device and 2.66x on H100 GPUs.
NVIDIA Releases GPU-Accelerated Robotics Guide Based on MjWarp
It presents migration procedures and performance tuning tips for scaling the SO-101 robot arm to up to 2,048 parallel GPU environments.
NVIDIA Releases Nemotron 3 Diarization
NVIDIA has released Nemotron 3 Diarization, a 100M parameter open-weight model that ranked first on the Voice Arena Diarization-Bench.
AISI and EvalEval Collaborate to Standardize LLM Evaluation Reproducibility
The UK's AISI and EvalEval have established a collaborative framework to enhance the reproducibility and verifiability of LLM evaluation results through the Every Eval Ever schema.
PickHugging Face Adds GGUF Model Support to Transformers
Hugging Face Transformers now directly supports the GGUF format from llama.cpp, simplifying local inference on Apple Silicon.
PickHugging Face Hires oMLX Developer to Expand MLX Support
Hugging Face has hired oMLX developer Jun Kim to strengthen support for the Apple Silicon-based MLX ecosystem and local AI infrastructure.
Multiverse Publishes Paper on LLM Block Removal Based on Ising Optimization
Multiverse Computing has unveiled a compression technique that transforms LLM block removal into an Ising optimization problem to minimize performance loss.
Hugging Face to Release tokenizers v1
Hugging Face plans to release tokenizers v1, aiming to reduce GPU idle time and achieve up to 30x faster processing speeds.
IBM Releases 'Consistency Gap' Metric for AI Agent Consistency
IBM Research defined the 'Consistency Gap' metric to measure the consistency of repeated AI agent executions and released a tool to improve it.