Tech Meridian ← ENTITY INDEX
PROMY MERIDIAN RU

COMPANY · ENTITY #11108

Hugging Face Hub

Related event timeline, sources and context from the news index.

EVENT TIMELINE

12

COMPANIES · 2 SOURCES · Hugging Face · The Decoder

Hugging Face routes language-model inference through OpenRouter

Hugging Face has switched its Inference Providers to OpenRouter: language-model requests from the Hugging Face Hub now go through OpenRouter’s router, which picks providers across 80+ partners based on live price, speed and reliability, retries failed calls, and bills usage to the user’s Hugging Face account while preserving the same API and token. Image, video and audio generation remain on Hugging Face’s existing providers; the change also exposes provider metrics, per-request billing breakdowns and controls to disable providers.

8.0

MODELS · 1 SOURCE · Hugging Face

Reson8 open-sources Rumble-2-Flash, a 0.5B multilingual ASR model

Reson8 released Rumble-2-Flash under the Apache 2.0 license: a 0.5-billion-parameter automatic speech recognition model supporting 40+ languages that the team says matches the accuracy of models 3–60x larger. It uses the Qwen3-ASR architecture with a GOLD-enabled on-policy distillation pipeline, runs in Hugging Face Transformers and vLLM without custom runtime changes, achieves over 7,000 RTFx on full batches, posts strong Open ASR Leaderboard and AppTek call-centre results (comparable to NVIDIA Nemotron 3.5 Transcribe on a 302-hour benchmark), and is available on the Hugging Face Hub and via Reson8's API.

8.0

COMPANIES · 1 SOURCE · Hugging Face

Hugging Face Hub billing can be routed through Google Cloud Marketplace

Hugging Face now allows organization admins to link their Hugging Face Hub organization to a Google Cloud billing account via the Google Cloud Marketplace so usage appears on the Google Cloud invoice. The integration covers Team/Enterprise subscriptions, pay-as-you-go compute (Jobs, Inference Endpoints, Inference Providers, GPU Spaces) and storage; subscriptions are reviewed by Hugging Face before billing is routed to Google Cloud.

5.0

MODELS · 1 SOURCE · Hugging Face

AnySpeechRecognizer and AnyDecisionModel let developers run open speech and decision models on Apple platforms

AnySpeechRecognizer and AnyDecisionModel are Swift packages that plug open speech-recognition and decision models into Apple’s speech API and local runtimes. They support Core ML-converted models (including NVIDIA Parakeet variants and Alibaba’s Qwen3‑ASR), speaker diarization via a Sortformer-based module, and on-device downloads from the Hugging Face Hub.

6.0

MODELS · 1 SOURCE · Hugging Face

Space lists which Transformers models run on AWS Trainium via torch-neuronx

A Hugging Face Space called the torch-neuronx model support Space publishes per-model test results and run scripts showing whether popular Transformers models run unmodified on AWS Trainium (Trainium2) using the torch-neuronx backend. The Space displays runtime stats, sharding and memory info, the Neuron stack (neuronx-cc) used, and exposes the underlying aws-neuron/torch-neuronx-model-support Parquet dataset and commands to reuse compiled graphs from the public NEFF cache bucket.

5.0

CODING · 1 SOURCE · AWS Machine Learning

Train Qwen3 on AWS Trainium using torchtitan

A torchtitan walkthrough demonstrates three ways to train Qwen3 on a single AWS Trainium trn2.48xlarge: (1) via Transformers v5 backend (AutoModelForCausalLM) reading/writing Hugging Face safetensors; (2) using torchtitan’s native Qwen3Model via config.model_spec; and (3) adding NKI (Neuron Kernel Interface) mixture-of-experts kernels from torchtitan-neuron and attaching custom kernels with the override registry. The post covers environment setup (Neuron PyTorch container or AWS DL AMI), installing torchtitan-neuron, torchrun launch patterns, checkpoint loading/saving (safetensors), and reports a verified step-1 loss example for Qwen3-4B-Instruct-2507 (6.54 from published weights vs 12.43 from random init).

5.0

CODING · 1 SOURCE · Hugging Face

Train in simulation and deploy walking policies to Pollen Robotics' Microduck with NVIDIA Isaac Lab 3.0

A newly published tutorial demonstrates how to train bipedal walking policies for the open-source Microduck robot using NVIDIA Isaac Lab 3.0 and Newton actuator models, export the policy as ONNX, and deploy it on the physical robot. The workflow uses GPU-parallel simulation, domain randomization, Newton/BAM actuator and backlash modeling, and automatic RSL-RL playback exports so models can be shared via the Hugging Face Hub or tested in Microduck Arena.

6.0

MODELS · 2 SOURCES · Hugging Face · The Decoder

Black Forest Labs releases FLUX.3 Image Klein with bounding-box layout control in Diffusers

Black Forest Labs published FLUX.3 Image Klein to the Hugging Face Hub and ships open checkpoints that run in Hugging Face Diffusers. The model supports text-to-image and multi-reference image-to-image generation, adds bounding-box layout control for region placement and image edits, includes a distilled checkpoint (1024×1024 ~2s on an H100, 16 steps by default), quantized single-file transformer weights (bf16, fp8r, nvfp4), and LoRA/DreamBooth training scripts.

8.0

MODELS · 1 SOURCE · Hugging Face

User automates fine-tuning and publishing of multiple small models using 'ML-intern' on Hugging Face Hub

A user reports using a conversational agent called ML-intern via HuggingChat to plan, budget, fine-tune, evaluate and publish several task-specific models and LoRAs on the Hugging Face Hub. Examples include a CPU-runnable 0.8B prompt-rewriter distilled from a 9B Qwen model, a Qwen3.5-2B fine-tuned 'citrus-disease-vlm-instruct' dataset model (base 14.9% → fine-tuned 52.8% on a 335-photo test; ~USD 1.90), and a Huggy LoRA trained on 84 drawings (compute ~USD 7.60); the author says the whole project’s compute including 8,797 example labels cost about USD 16.

6.0

MODELS · 1 SOURCE · AWS Machine Learning

Z.ai GLM 5.3 (753B MoE) now available on Amazon Bedrock

Z.ai’s GLM 5.3, a 753‑billion‑parameter mixture‑of‑experts model optimized for coding and long‑horizon agentic tasks and reported to show strong cybersecurity performance, is now available to eligible enterprise customers via Amazon Bedrock. The Bedrock integration offers OpenAI‑compatible APIs (Responses and Chat Completions), Bedrock Invoke/Converse, cross‑Region inference, implicit/explicit prompt caching, and service tiers to simplify managed access and reduce latency and cost.

7.0

MODELS · 1 SOURCE · Hugging Face

Open TTS Leaderboard launches objective, scalable evaluation for multilingual TTS and voice cloning

The Open TTS Leaderboard is an open evaluation platform that ranks TTS models using objective metrics—intelligibility (WER/CER via Qwen3 ASR), speaker similarity (cosine SIM via WavLM), and speed (RTFx and TTFA measured on H200/CPU)—across Seed TTS Eval and CV3 Eval datasets. It supports multilingual and voice-cloning comparisons, provides Pareto visualizations and a Listen tab to audition outputs, and aims to speed evaluation from weeks to hours.

6.0

COMPANIES · 1 SOURCE · Hugging Face

Hugging Face Hub adds dedicated RL Environments dataset filter and framework tags

Hugging Face Hub introduced a dedicated spot for reinforcement-learning environments: dataset repositories tagged rl-environment now appear in a new filter and show generated "Use this dataset" commands plus framework compatibility tags (e.g., Harbor, Verifiers, NVIDIA NeMo Gym, OpenEnv). The release focuses on publishing tasksets (tasks, tests, verifiers) in repo datasets to reduce siloed environment formats and to enable running and verifying tasks via Harbor, NeMo Gym, and Hugging Face Jobs/Sandboxes.

6.0