cohere

This page hosts GGUF quantized versions of Cohere's North-Mini-Code-1.0 model, a 30B-A3B MoE model optimized for code generation and agentic tasks. Instructions are provided for building llama.cpp from a specific PR to support the cohere2moe architecture.

Cohere officially launches North Mini Code, a coding model, with weights available on Hugging Face and deployment support for vLLM and MLX.

Cohere released North Mini Code, its first open-source coding model under Apache 2.0, designed to be small, cost-effective, locally deployable, and focused on agentic performance.

Cohere and Cohere Labs released North Mini Code, an open weights 30B-A3B parameter model optimized for code generation, agentic software engineering, and terminal tasks, with strong benchmark results on SWE-Bench and Terminal-Bench.

Cohere released its first open-source coding model, North Mini Code Small, designed for efficient agentic performance and community input.

Cohere released North Mini Code, a 30B-parameter Mixture-of-Experts model with 3B active parameters under Apache 2.0, optimized for agentic software engineering tasks and outperforming similar-sized models on coding benchmarks.

Cohere has released an early access coding model, BLS-Mini-Code-1.0, a 30B parameter model available on Hugging Face for testing.

Cohere Labs released North Mini Code, a 30B-parameter (3B active) open-weights model optimized for code generation, agentic software engineering, and terminal tasks, licensed under Apache 2.0.

A PR for mlx-lm adds support for Cohere's Command A+ (218B MoE) model on Apple Silicon, with architecture details for the implementation.

Cohere released Command A+, a 218B mixture-of-experts model with 25B active parameters, consolidating five previous models into one with improved performance on enterprise agentic tasks. It is open-sourced under Apache 2.0 and available on Hugging Face.

Fine-tuned Cohere Transcribe, the best open-source speech-to-text model, to support diarization and timestamps. The new model is available on Hugging Face.

Cohere has released Command A+, an open enterprise AI model designed for business applications.

Cohere releases Command A+, a 218-billion-parameter sparse MoE language model under the permissive Apache 2.0 license, featuring lossless 4-bit quantization and native citations for enterprise and sovereign AI use.

Cohere launches Command A+, its first Mixture-of-Experts model, released under Apache 2.0 with efficient quantization for 1-2 GPU deployment, prioritizing practicality and open access for developers.

Cohere announces Command A+, their most powerful open-source LLM yet, optimized for efficiency and released under the Apache license.

Cohere has released Command A+, an open-source model with 25 billion active parameters and 218B total parameters under Apache 2.0, optimized for agentic, multilingual, and reasoning-heavy tasks.

Cohere releases Command A+, an open-source model with 25B active parameters (218B total) optimized for agentic, multilingual, and reasoning-heavy tasks, supporting vision inputs and 128K context under Apache 2.0.

Cohere has released Command A+, its most powerful open-source LLM, optimized to run on minimal hardware.