DeepSeek

DeepSeek's documented successor to the V3.2 experimental line, positioned in official DeepSeek API news as live on app, web, and API.

Microsoft AI

Microsoft AI's faster MAI image variant, announced for lower-latency image generation and editing workflows.

OpenAI

OpenAI's smallest GPT-5.4 nano model, documented in the official OpenAI API model guide for very low-latency or economical GPT-5 family routing.

OpenAI

Whisper large-v3 is OpenAI’s ASR model for transcription and translation across many languages, with strong robustness to accents and noise. It is commonly self-hosted or used via API partners; latency depends heavily on hardware and chunking strategy.

OpenAI

text-embedding-3-large produces high-dimensional text embeddings for semantic search, clustering, and classification. Teams pair it with pgvector or SaaS vector DBs for RAG; output dimensions can be reduced with tradeoffs described in OpenAI documentation.

Snowflake

Snowflake Arctic is an enterprise-oriented open model emphasizing efficient training recipes and SQL-adjacent enterprise tasks inside the Snowflake ecosystem. It targets teams that want LLM features colocated with governed data in Snowflake Cortex.

Microsoft

Phi-4 is Microsoft Research’s small language model line focused on strong reasoning per parameter for on-device and low-cost cloud scenarios. Deployment often happens via Azure AI or Hugging Face hubs—confirm license for your channel.

OpenAI

Compact reasoning-focused model in OpenAI’s o-series line aimed at strong STEM and coding performance with lower cost than full o3. Intended for developers who want reasoning without always paying flagship prices—confirm exact API availability and snapshot names in OpenAI docs.

NVIDIA

NVIDIA Nemotron-4 340B is a large open-weights model suite aimed at enterprise and research users who train and serve on NVIDIA stacks (NeMo, NGC). It targets GPU-native teams that need customizable checkpoints with NVIDIA-optimized tooling.

Mistral AI

Mistral Small 3 is Mistral’s efficiency tier for fast, affordable chat and tool use at high QPS—positioned between tiny open models and Mistral Large. Exact naming and versioning appear in Mistral’s API catalog; pin versions in production.

Mistral AI

Mistral 7B Instruct is a compact dense model that popularized efficient open-weight chat quality at small scale. It remains a baseline for fine-tunes and on-prem pilots where 13B+ models are too heavy.

Meta

Llama 3.1 8B Instruct is a small open-weights model for edge laptops, single-GPU servers, and ultra-low-latency assistants. Quality per dollar is competitive for simple tasks but not for frontier reasoning.

Meta

Llama 3.1 70B Instruct is a mid-size open-weights instruct model balancing quality and deployability on a single large GPU or small multi-GPU nodes. Common for private assistants, on-prem pilots, and fine-tunes where 405B is impractical.

Google

Gemma 2 27B is Google’s open-weights Gemma family checkpoint balancing quality and deployability for research and product teams that need permissive terms without Vertex-only APIs. It is often fine-tuned for domain tasks on TPU or GPU clusters.

Google

Gemini 2.0 Flash is Google’s efficiency-oriented multimodal model generation aimed at fast agentic and interactive experiences. Capabilities and naming evolve—validate against the current Gemini API reference for tool use and context limits.

AWS

Titan Text Premier is AWS’s managed text model for Bedrock workloads emphasizing integration with guardrails, knowledge bases, and private data patterns. It targets enterprise RAG and internal assistants rather than frontier creative writing.

Mistral AI

Catalog entry for this named release; see the provider’s official documentation for modalities, pricing, and context limits.

Community

Catalog entry for this named release; see the provider’s official documentation for modalities, pricing, and context limits.

xAI

Catalog entry for this named release; see the provider’s official documentation for modalities, pricing, and context limits.

DeepSeek

Catalog entry for this named release; see the provider’s official documentation for modalities, pricing, and context limits.

Databricks

Catalog entry for this named release; see the provider’s official documentation for modalities, pricing, and context limits.

Anthropic

Catalog entry for this named release; see the provider’s official documentation for modalities, pricing, and context limits.

Microsoft AI

Microsoft AI's frontier reasoning model in the MAI family, announced for difficult prompts, science, math, and complex planning workloads, with Microsoft Foundry access documented as private preview.

Anthropic

Anthropic's experimental Claude model for open-ended scientific inquiry and long-horizon reasoning, documented as limited to approved Project Glasswing researchers.