Coding playbooks · Reference table
Model specs: context, tools, and cutoffs in one table
Written November 6, 2026
Context windows, tool support, modalities, and knowledge cutoffs for the models coding agents actually use — each cell linked to its source.
Every cell below traces to the vendor's own model page or model card. A dash means the vendor page does not state that value. It is a gap in the vendor's page, not a number filled in from a second-hand summary. Specs move at release time, so open the linked source before planning a workload around a row.
How to read this table
A context window is the total budget the model works inside, not the amount you can usefully fill. Input, thinking, and output can share one window: Gemini 4 Argon counts all three against the same 1,048,576 tokens, while some vendors split the window between input and output instead. A window is also not the same as usable context: the advertised number is the ceiling, and hosts and harnesses can cap it far below. The Ripple-65B card advertises 1,048,576 tokens, but the hosted endpoint enforces a 16,384-token limit, per our Ripple-65B page. Check the live route, not just the card, before planning long-context work. The table shows specs, not a ranking. A bigger window says nothing about whether the model does the job well.
The table
| Model | Context window | Max output | Tools | Modalities | Knowledge cutoff | Source |
|---|---|---|---|---|---|---|
| GPT-6.1 Sol | 1.05M tokens | 128K tokens | Functions, web search, file search, computer use | Text, image in; text out | Apr 30, 2026 | OpenAI API models |
| GPT-6.1 Astra | 1.05M tokens | 128K tokens | Functions, web search, file search, computer use | Text, image in; text out | Apr 30, 2026 | OpenAI API models |
| Gemini 4 Argon | 1,048,576 tokens (unified input + output) | 65,536 tokens by default; up to 1,048,576 with long decode | Function calling, code execution, file search, computer use | Text, image, video, audio, PDF in; text out | Mar 2026 | Gemini 4 Argon model page |
| Kimi K3.1 Pro | 1,048,576 tokens | — | Tool calls, tool choice | Text, image | — | Kimi K3.1 Pro model card |
| Mistral Large 4 | 1M tokens | — | Native function calling, JSON output | Text, image in; text out | — | Mistral Large 4 model card |
| LongCat-2.5 | — | 256K tokens | Function calling | Text, image | — | LongCat API FAQ |
| Ripple-65B | 1,048,576 tokens | — | Yes; MCP Atlas score on the model card | Text | — | Ripple-65B Preview model card |
| Hy4 | 1M tokens | — | Tool-call parser in the serving recipe | Text | — | Hy4 preview model card |
| Claude Sonnet 5.5 | 1M tokens | 128K tokens | Agentic tool use | Text, image | Jun 2026 | Claude Sonnet 5.5 overview |
| Claude Opus 5.5 | 1M tokens | 128K tokens | Agentic tool use | Text, image | Jun 2026 | Claude Opus 5.5 overview |
| GPT-6 Cyber | 1,050,000 tokens | 128,000 tokens | Function calling; web search, file search, code interpreter, MCP, computer use | Text, image in; text out | — | GPT-6-Cyber model page |
What this table does not do
It does not rank models by context window alone, and it does not tell you what a model costs. For that, see Model prices: what the major models cost and, if you pay through a subscription instead of the API, Plan allowances: what each agent plan lets you do. It does not cover licences: open-weight rows here are described by their specs only, and the licence terms live in Open-weight model licences, in one list. A dash is a missing vendor statement, not a missing capability. Where a vendor does not publish a value, the honest entry is blank. And a context window is a ceiling on the vendor's page, not a measurement of how much of it your setup will actually let you use. Some agent setups assume a smaller window for models they do not recognize and compact early, so confirm the working window in your own setup before designing a long session around a row.
Related tools
- Model prices: what the major models cost
- Open-weight model licences, in one list
- What broke in agent tools this week
- Agent change deadlines: retirements and shutdowns
Sources
- Models — OpenAI API. GPT-6.1 Sol and GPT-6.1 Astra context, output, tools, modalities, and cutoffs
- GPT-6-Cyber Model — OpenAI API. GPT-6 Cyber context, output, function calling, tools, and modalities
- Gemini 4 Argon — Google AI for Developers. Argon context window, output limits, function calling, and input types
- Gemini 4 Argon — Model Card — Google DeepMind. Argon knowledge cutoff, March 2026
- Kimi K3.1 Pro model card — Hugging Face. Kimi context length, modalities, and tool calls
- Mistral Large 4 model card — Hugging Face. Mistral context length, modalities, and native function calling
- FAQ — LongCat API Docs. LongCat-2.5 maximum output, 256K tokens
- Responses API — LongCat API Docs. LongCat-2.5 function calling and image input
- Change Log — LongCat API Docs. LongCat-2.5 native multimodality and tool use
- Ripple-65B Preview model card — Hugging Face. Ripple context length and text-only pretraining
- Hy4 preview model card — Hugging Face. Hy4 context length and tool-call parser in the serving recipe
- Claude Sonnet 5.5 — Claude Platform Docs. Sonnet 5.5 context, output, cutoff, visual understanding, and agentic tool use
- Claude Opus 5.5 — Claude Platform Docs. Opus 5.5 context, output, cutoff, vision-heavy workflows, and agentic tool use
Back to: Coding playbooks hub · Also: Model prices: what the major models cost · Plan allowances: what each agent plan lets you do · Open-weight model licences, in one list