Agent Stack Landscape

Graph-based orchestration runtime for long-running, stateful agents.

Framework for building LLM-powered applications and workflows.

AI-native code editor with codebase context, multi-file agents, Projects (persistent coordinators), Origin code hosting, self-hosted Cloud Agent machines, and intelligent model routing for teams.

Data framework for connecting LLMs to private and structured data.

OpenAI develops the GPT family of foundation models, ChatGPT, developer APIs, and enterprise AI platforms. It helped popularize large language models through ChatGPT and now powers AI assistants, coding tools, enterprise copilots, retrieval systems, and agent-based applications across industries. Its Decisions API (`POST /v1/decisions`) became generally available on 2026-10-22 for `gpt-6-luna`, returning probabilities for yes/no, multiple-choice, and score questions, with no output-token charge. GPT-6.1 Terra (`gpt-6.1-terra`, 2026-10-26) is the current GPT-6 Terra for visual reasoning and computer use at $1/$5 Standard. As of 2026-10-21, GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Luna are in private preview in Google Cloud Model Garden for select enterprise customers; general availability is planned in the coming weeks, and Microsoft remains the primary cloud partner.

Anthropic builds the Claude family of foundation models for enterprises that prioritize instruction following, long-context reasoning, coding quality, and predictable, safety-conscious behavior. Constitutional AI and careful system design are core to how Claude is trained and positioned. Claude Opus 5.5 (claude-opus-5-5) shipped 2026-09-22 at $4/$20 per 1M tokens, Sonnet 5.5 on 2026-09-28, and Haiku 5.5 (claude-haiku-5-5) on 2026-10-07 from $0.10/$0.50, completing the 5.5 lineup. As of 2026-09-16, Claude apps merge chat and Cowork into one conversation (Pro/Max first), with Claude Docs, Slides, and Design available from any chat; on 2026-10-08 those three left beta for every plan, including Free, and Claude Dashboards (live dashboards over connected data platforms, paid plans) and Claude Motion (code-built animations, Team and Enterprise) entered beta. Claude Code Projects (beta, 2026-09-17) add a coordinator that delegates to parallel cloud threads. On 2026-09-18 Anthropic announced an embedded-evaluation partnership with Accenture’s Faculty unit. As of 2026-10-22, Claude Models on Google Cloud (Anthropic managed) reaches the Claude API through a Google Cloud project: IAM controls access and Marketplace handles billing. Fast mode is available on that surface for Claude Opus 5.5.

Google combines frontier AI research (DeepMind), cloud infrastructure (Vertex AI), custom AI hardware (TPUs), and productivity products (Workspace, Android, Search), making it one of the most vertically integrated AI platforms. Gemini is the model layer; Google Cloud and consumer surfaces are the distribution layer. Gemini 4 Argon (gemini-4-argon, 2026-10-13) is the public flagship on the paid Gemini API. Gemini 4 Carbon was announced 2026-10-28 for cyber defenders and then Fairwind partners; Google has not published a public API id. Genie 4 (2026-10-27) is a research-preview world model for trusted testers, not a Gemini chat id. Gemini 3.8 Flash remains the lower-priced workhorse. Gemini 3.8 Live (2026-09-15) is the Live API voice SKU, with Live Avatar (2026-09-24) in Gemini Enterprise. Gemini 3.8 Flash TTS (2026-09-23) is a separate speech-generation model, distinct from 3.8 Flash chat. As of 2026-10-21, GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Luna are in private preview in Model Garden for select enterprise customers.

Meta advances open multimodal research and releases the Llama open-weight model family used widely in self-hosted AI stacks. Muse Wonder (muse-wonder-1.0, 2026-10-12) is the newer paid Meta Model API peak for longer agentic work, with Muse Code. Muse Spark 1.3 remains the lower-priced closed API. Muse Spark Open (2026-10-15) is a separate weight release: 816B total / 42B active, 1M context, multimodal, under the Muse Model License with limited commercial terms. Combined with PyTorch, Llama still lets organizations train, fine-tune, and serve models without depending on a proprietary API.

Microsoft delivers Azure AI, Microsoft 365 Copilot, GitHub Copilot, Phi small models, and Semantic Kernel—plus deep distribution of OpenAI models through Azure OpenAI Service and Microsoft Foundry (GPT-6 Astra as of 2026-09-03). It is the default enterprise path for copilots, governed LLM APIs, and AI that sits inside Office, GitHub, and Azure workloads.

Databricks builds the Lakehouse platform and Mosaic AI for training, serving, and governing models on enterprise data—plus open projects like MLflow, Delta Lake, and Unity Catalog. Teams use it when RAG, agents, and fine-tuning need to run next to governed tables. Kimi K3 is available through Unity AI Gateway (2026-08-06) alongside proprietary frontier APIs.

Snowflake runs the Data Cloud for analytics and increasingly AI—Cortex functions, Arctic models, and LLM apps that stay close to governed tables. Grok 4.6 is in public preview on Cortex AI Functions and Cortex Inference as of 2026-09-03. GPT-6 Astra is in private preview on Cortex AI as of 2026-09-09.