Catalog of supported open-weight models across families including Meta (Llama ecosystem), Alibaba Cloud Qwen, Mistral AI, DeepSeek, NVIDIA Nemotron, IBM Granite, Nomic, and Cobble-built pipelines. Pricing assumes reclaimed GPU infrastructure, efficient vLLM serving, and open-weight licensing—benchmarked against typical marketplace rates.
Labels such as Flagship, Enterprise ready, Multilingual, and Best for RAG appear as tags on each card.
Featured
Cobble's flagship reasoning model offering frontier-class performance with Mixture-of-Experts efficiency, ideal for research, coding, and complex agent workflows.
$0.23 / 1M input tokens · $1.87 / 1M output tokens · $0.4675 / 1M cached tokens
Featured
Cobble's document extraction pipeline for archival and research-grade text processing.
$4 / 1K pages
Featured
State-of-the-art open embedding model with strong multilingual and long-context support.
$0.01 / 1M input tokens
Alibaba Cloud
Cobble's flagship reasoning model offering frontier-class performance with Mixture-of-Experts efficiency, ideal for research, coding, and complex agent workflows.
Alibaba Cloud
Balanced high-performance model with strong reasoning and code generation at lower latency and cost.
Alibaba Cloud
Sparse MoE architecture delivering high quality responses with excellent cost efficiency.
Google DeepMind
Large open model with excellent instruction following, multilingual capabilities, and coding performance.
Alibaba Cloud
Low-latency utility model ideal for chatbots, summarization, and lightweight automation.
Google DeepMind
Efficient sparse variant of Gemma optimized for strong quality with lower serving costs.
Google DeepMind
Unified encoder-free multimodal model handling text, image, audio, and video with strong quality at small-model cost.
Google DeepMind
Efficient 4B-class model tuned for high-volume, low-latency workloads like classification and extraction.
Google DeepMind
Ultra-light 2B-class model for massive-scale pipelines, routing, and lightweight chat at minimal cost.
Mistral AI
Compact Mistral 3-generation model with strong function calling and multilingual chat at flat token pricing.
Mistral AI
Smallest Mistral 3-generation model, ideal for edge-style automation, drafting, and structured output.
Mistral AI
Beloved creative workhorse with natural prose, strong multilingual range, and dependable instruction following.
OpenAI
OpenAI's open-weight MoE model (3.6B active) with strong reasoning and tool use at very low cost.
DeepSeek
Efficiency-optimized Mixture-of-Experts model (284B total / 13B active) built for fast inference over a 1M-token context window.
DeepReinforce AI
Compact self-scaffolding coding model from the Ornith 1.0 family, punching far above its size on agentic coding benchmarks.
DeepReinforce AI
Sparse MoE coding model (~3B active per token) that rivals much larger frontier models on Terminal-Bench and SWE-Bench.
DeepReinforce AI
Dense 31B variant of the Ornith 1.0 agentic coding family, MIT-licensed with strong tool-use performance.
Zhipu AI
General-purpose OCR model with strong support for complex layouts and multilingual documents.
DeepSeek
High-accuracy OCR and document understanding model optimized for tables and technical PDFs.
NVIDIA
Enterprise-grade OCR for forms, scanned documents, and large ingestion pipelines.
Cobble Labs
Cobble's document extraction pipeline for archival and research-grade text processing.
IBM
High-quality enterprise embedding model for semantic search and RAG applications.
IBM
Fast, lightweight embedding model for large-scale indexing and low-cost retrieval.
Nomic AI
State-of-the-art open embedding model with strong multilingual and long-context support.
Alibaba Cloud
Compact multilingual embedding model offering strong performance and low cost.
Alibaba Cloud
Flagship embedding model with excellent retrieval accuracy across multilingual corpora.
