The Open Weights
LatestModelsLeaderboardsCompanies
Subscribe
The Open Weights

The daily record of open-source AI. New model releases, leaderboards, and what's coming next — written for people who ship.

Refreshed every 12 hours

Discover

  • Latest releases
  • New today
  • Trending models

Browse

  • All models
  • Companies
  • Categories
  • Leaderboards

About

  • About
  • Editorial policy
  • Hardware estimates
  • RSS feed
  • llms.txt
  • Newsletter

© 2026 The Open Weights. An independent publication.

PrivacyTermsSMSAggregated by Claude · curated by humans.
LatestCohere1.0
CohereVision-Language

Cohere Labs releases compact North Micro Vision model

A small multilingual vision-language model built for instruction following arrives under a research-only license.

Aug 10, 2026
NotableCC BY-NC 4.0
North Micro Vision Instruct

Cohere Labs has published North Micro Vision Instruct, a compact vision-language model tuned to follow instructions across text and images. The release marks the initial 1.0 version of the North Micro Vision family and lands on Hugging Face for researchers to evaluate.

The model is positioned as a small, multilingual entry in the increasingly crowded field of vision-language models (VLMs). Rather than chasing the largest possible parameter counts, North Micro Vision emphasizes a lighter footprint aimed at applications where multilingual understanding and image comprehension matter more than sheer scale.

Why it matters

Compact VLMs are useful precisely because they lower the cost of running multimodal systems, making them practical for teams that cannot dedicate large amounts of compute to inference. A multilingual focus also broadens the model's reach beyond English-centric benchmarks, which remains a persistent gap in many open releases.

A few things to keep in mind about this release:

  • It ships under a CC-BY-NC-4.0 license, meaning use is restricted to non-commercial and research settings.
  • It is an instruct-tuned model, oriented toward following prompts rather than serving as a raw base checkpoint.
  • Cohere Labs has not published detailed parameter counts or context-length specifications alongside the release.

The non-commercial license places North Micro Vision firmly in the research lane, in keeping with much of Cohere Labs' open work. Teams interested in the model's exact capabilities should consult the model card on Hugging Face as further documentation and benchmarks become available.

Sources

  • CohereLabs/North-Micro-Vision-Instruct

    Hugging Face

    Visit
OlderVak Conformer targets speech recognition in six Indic languagesShunyalabs · Speech → Text · 3 months agoNewerNVIDIA Opens Magpie TTS for Multilingual Voice AgentsNVIDIA · Text → Speech · 3 months ago

Get the model

Hugging Face

Specs

Context window500K tokens
Size5 GB
PrecisionBF16
ArchitectureCohereCompassForConditionalGeneration
LicenseCC-BY-NC-4.0
Downloads16.3K
Likes155

Can you run it?

Runs on just about anything.

  • BF16 (as published)

    8 GB GPU (RTX 4060) · 16 GB laptop or Mac

    6.9 GB
  • 8-bit

    8 GB GPU (RTX 4060) · 16 GB laptop or Mac

    4.6 GB
  • 4-bit

    Any modern laptop · 8 GB RAM

    3.4 GB
Your machine
  • KV cache sized for 8,192 tokens of context; longer prompts need proportionally more.

Based on Hugging Face weights metadata. Assumes an 8K context and ~1 GB runtime overhead; actual needs vary. How we estimate


Modalities

Vision-Language

The Weekly Weights

Every open release that mattered, one email a week.

0 comments

No comments yet. Be the first to weigh in.

More from Cohere

All Cohere releases →
Cohere Transcribe Arabic
Cohere/Speech → TextRuns anywhere

Cohere releases Apache-licensed Arabic speech model

The Cohere Labs transcription model targets Arabic and English audio under a permissive open license.

Jun 18, 2026
North-Mini-Code 1.0
Cohere/CodeConsumer GPU

Cohere Releases North-Mini-Code, an Open MoE Model

The new Apache 2.0-licensed model is designed for code generation and agentic chat applications, using a Mixture-of-Experts architecture for efficiency.

Jun 5, 2026
Cohere Transcribe 03-2026
Cohere/Speech → TextRuns anywhere

Cohere Releases Top-Ranked Multilingual Transcription Model

The new automatic speech recognition model from Cohere Labs sets a new benchmark on the Hugging Face Open ASR Leaderboard for multilingual performance.

Mar 24, 2026

More in Vision-Language

All Vision-Language →
SenseNova-Loop-9B
SenseTime/Vision-LanguageRuns on a laptop

SenseTime Releases Looped 9B Multimodal Model

The company's new SenseNova-Loop-9B pairs vision-language capabilities with a recurrent looped-transformer design.

Nov 17, 2026
Hunyuan Hy4-flash
Tencent/Vision-LanguageMulti-GPU server

Tencent releases Hunyuan Hy4-flash multimodal MoE

The lightweight, mixture-of-experts model adds vision and text instruction-following to Tencent's open Hunyuan lineup.

Nov 17, 2026
Mirage
Cactus Compute/Vision-Language

Cactus Compute's Mirage brings vision tasks on-device

The new open vision-language model targets detection, segmentation, and visual Q&A without relying on the cloud.

Nov 5, 2026