tensorfolk

Microsoft's MAI Image 2.7 preview ranks second for image editing on Artificial Analysis

Microsoft AI released MAI Image 2.7 in preview as an editing-first model that can restyle or relight scenes, change text inside images and restore old assets. It debuts at number two on Artificial Analysis's editing leaderboard and number three on Arena's edit board. Editing is open to try on those sites and MAI Playground; text-to-image generation and API access have not shipped yet.

Why it matters

Another strong option for instruction-based photo edits, but you can't build on it until an API appears.

Source: Read the full story on Microsoft AI

microsoftimage-editingmaileaderboards

More in Generative Media

Generative Media / Image Generation

Canva launches Craft, its own image model built for layouts and legible text

Canva Research released Craft, which plans a layout before rendering images up to 2K in many aspect ratios. Trained largely on Canva's design templates, it targets readable text and partial edits. It is live in Dream Lab for all users and through the Canva API, with faster versions planned.

Why it matters: A design-focused model aimed squarely at text rendering and layout, where general image models often struggle.

Read on BetaNews
Generative Media / Voice, Speech & Voice Agents

NVIDIA opens Nemotron 3.5 Transcribe, a 30B-A4B speech model for 40 languages, plus 463k-hour Granary v2

NVIDIA published open weights for Nemotron 3.5 Transcribe, a mixture-of-experts speech model with about 4B active parameters. One model handles transcription in 40 languages, translation to and from English for 31 of them, and labeling of up to eight speakers in hour-long recordings. Alongside it comes Granary v2, a speech corpus covering 48 languages.

Why it matters: One commercially usable open model replaces separate ASR, translation and diarization pipelines, and it tops the Open ASR leaderboard among open-weight systems.

How to try it

Nemotron 3.5 Transcribe is an open-weight (OpenMDW-1.1) speech model you can run yourself for transcription, translation and speaker-labelled transcripts of recorded audio. NVIDIA's reference setup is a single H100 80GB.

  1. Set up the NeMo Speech environment as described on the model card.
  2. Download the checkpoint from Hugging Face (nvidia/NVIDIA-Nemotron-3.5-Transcribe-30B-A4B).
  3. Load it with NeMo's ASRModel.from_pretrained and call transcribe() on a WAV file.
  4. For serving, follow the model card's vllm serve command and send audio through the OpenAI-compatible chat API.
python
import nemo.collections.asr as nemo_asr
asr_model = nemo_asr.models.ASRModel.from_pretrained("nvidia/NVIDIA-Nemotron-3.5-Transcribe-30B-A4B")
transcriptions = asr_model.transcribe(["file.wav"])
Open the docs
Big newsRead on NVIDIA (Hugging Face blog)
Generative Media / Voice, Speech & Voice Agents

Rime launches Prism, a fine-tunable speech-to-speech model that generates text before audio

Voice startup Rime released Prism, an enterprise speech-to-speech model that companies can fine-tune and run on-premises or in a private cloud. Because the model produces text before it speaks, businesses can check each reply against their own rules and keep audit logs. Rime also says it detects caller emotions like frustration, and Dialpad was a design partner.

Why it matters: Text-first generation offers a practical way to add guardrails and auditability to voice agents in regulated settings like contact centers.

Read on Rime (PR Newswire)