LLMs with largest context windows
Read Time 15 mins | Written by: Cole
[Last updated: July 2026]
The largest LLMs today support context windows ranging from 400K to 1 million input tokens—enough to ingest entire codebases, hundreds of legal contracts, full video transcripts, or months of agent session history in a single pass. A million tokens is roughly 750,000 words, or the equivalent of 10–15 full-length novels processed at once.
That scale unlocks a new tier of practical use cases: coding across multi-repo projects without chunking, agentic workflows that hold full tool-call history in context, long-term memory for AI assistants, end-to-end analysis of legal or financial document sets, and multimodal pipelines combining text, images, and video in one request.
The engineering workarounds that used to define long-context work—RAG, sliding windows, lossy summarization—are increasingly optional (but still important).
LLMs with largest context windows:
- 100M tokens: Magic.dev’s LTM-2-Mini enables processing of enormous datasets – entire code repositories (up to 10 million lines of code) or large-scale document collections (750 novels). We still haven’t seen evidence anyone is using this model or its 100 million token context.
- 10M tokens: Meta’s Llama 4 Scout runs on a single GPU, perfect for on-device multimodal workflows, deep video/audio transcript analysis, and full-book summarization.
- 1M tokens: Claude Fable 5, Claude Opus 5, Claude Sonnet 5, OpenAI's GPT-5.6 Sol, Terra, and Luna, Google's Gemini 3.6 Flash and Gemini 3.5 Flash, Moonshot's Kimi K3, DeepSeek V4, and Meta's Llama 4 Maverick are the current frontier for complex multimodal tasks, enterprise-grade document analysis, and large-scale codebase comprehension. Anthropic ships 1M at standard pricing with no long-context surcharge, and DeepSeek made 1M the floor for every model in its V4 lineup.
- ~984K tokens: Qwen 3.8 is Alibaba's 2.4 trillion parameter sparse MoE model, with a 131K output window.
- 500K tokens: xAI's Grok 4.5 is the one frontier flagship that did not follow the field to 1M, trading window size for speed and price at $2/$6 per million tokens.
- 400K tokens: OpenAI’s GPT-5.4 mini and GPT-5.4 nano deliver extended context at lower cost, with a large 128K output window and strong long-context performance for agentic tasks.
- 256K tokens: Mistral Large 3 and Mistral Small 4 are the strongest open-weight options for regulated and on-premises deployments. Small 4 folds instruction following, reasoning, vision, and agentic coding into a single 119B model.
Let’s take a closer look at what these models can do with their large context windows.
Model details for LLMs with large context windows
Input Context Window – Up to 100 million tokens
Magic.dev's LTM-2-Mini boasts an extraordinary 100 million token (10 million lines of code or 750 novels) context window, making it the largest context window available. This model is built for handling massive datasets, like entire codebases or vast collections of documents.
Primary Use Cases
- Ultra-long codebase comprehension and refactoring
- Legal-contract and policy analysis spanning thousands of pages
- Full-book summarization and knowledge extraction
We still haven’t seen evidence anyone outside of Magic.dev is using this model or its 100 million token context.
Input Context Window – Up to 10 million tokens
A 109B parameter MoE model with 17B active parameters and 16 experts, Scout delivers an unprecedented 10 million-token window on a single NVIDIA H100 GPU. It outperforms competitors like Google's Gemma 3 and Mistral 3.1 across benchmarks while supporting native multimodality.
Primary Use Cases
- On-device multimodal workflows requiring ultra-long context
- Large-scale codebase comprehension and automated refactoring
- Full-book summarization and deep video/audio transcript analysis
Input Context Window – 1 million tokens
Claude Fable 5 is Anthropic's most capable widely released model, generally available since June 9, 2026 at $10/$50 per million input/output tokens. It runs adaptive thinking always-on and supports 128K max output.
Anthropic also ships Claude Mythos 5, which shares Fable 5's specs and pricing but is invitation-only through Project Glasswing for defensive cybersecurity work. There is no self-serve access.
Primary Use Cases
- Long-running autonomous agents that need to hold days of session history in context
- Frontier-level coding and knowledge work where capability matters more than cost
- Complex multi-document synthesis across research, legal, and financial corpora
Input Context Window – 1 million tokens
Claude Opus 5 is Anthropic's model for complex agentic coding and enterprise work, priced at $5/$25 per million input/output tokens with the full 1M context window included at standard rates. A 900K-token request costs the same per-token rate as a 9K one.
Supports 128K max output on the Messages API, or up to 300K output tokens through the Message Batches API with the extended-output beta header. Available on the Claude API, Amazon Bedrock, Claude Platform on AWS, Google Cloud, and Microsoft Foundry.
Primary Use Cases
- Large-scale codebase comprehension and multi-file refactoring without chunking
- Enterprise contract, legal, and research document analysis across hundreds of pages
- Long-running agentic workflows where full session history, tool calls, and intermediate reasoning stay intact
Input Context Window – 1 million tokens
Claude Sonnet 5 delivers the best combination of speed and intelligence in the Claude lineup, with the same 1M context window and 128K output ceiling as Opus 5.
Standard pricing is $3/$15 per million input/output tokens. Introductory pricing of $2/$10 applies through August 31, 2026, which makes it the cheapest way to run 1M-token workloads on Claude until then.
Primary Use Cases
- High-throughput document processing and synthesis
- Production developer applications requiring large codebase context
- Cost-efficient alternative to Opus 5 for long-context use cases at scale
OpenAI GPT-5.6 Sol, Terra, and Luna
Input Context Window – 1 million tokens
Released July 9, 2026, the GPT-5.6 family replaced GPT-5.5 across ChatGPT, Codex, and the API. Sol is the flagship at $5/$30 per million input/output tokens, Terra is the balanced tier at $2.50/$15, and Luna is the cost-efficient tier at $1/$6.
OpenAI's own long-context benchmarks show Sol at 73.8% on MRCR v2 8-needle between 512K and 1M tokens, and 77.1% on GraphWalks BFS at 1 million tokens. The ultra setting coordinates four agents in parallel by default, trading token spend for faster results on demanding tasks.
Programmatic Tool Calling in the Responses API lets the model write and run in-memory programs that filter large intermediate results, which cuts token use on tool-heavy work rather than pushing everything back through the context window.
Primary Use Cases
- Agentic workflows requiring computer use and parallel subagents
- Large-scale code generation and enterprise multi-step automation
- Tool-heavy pipelines where intermediate data can be filtered programmatically
Google Gemini 3.6 Flash and Gemini 3.5 Flash
Input Context Window – Up to 1 million tokens
Gemini 3.6 Flash, stable as of July 21, 2026, is Google's latest model balancing speed with intelligence on agentic and multimodal tasks. Gemini 3.5 Flash sits alongside it as the most intelligent option for sustained frontier performance on agentic and coding work.
Worth knowing before you architect around it: Gemini 3.1 Pro is still Preview status, not stable, and Gemini 3 Pro Preview has been shut down. For production workloads, the Flash line is where Google's stable long-context capacity actually lives.
Primary Use Cases
- Complex multimodal workflows (video, audio, images, and text in one shot)
- Advanced coding assistants and autonomous agent pipelines
- Semantic search and enterprise document analysis at scale
Input Context Window – Up to 1 million tokens
Kimi K3 launched July 16, 2026 as a 2.8 trillion parameter mixture-of-experts model with native multimodal support, activating a small fraction of its experts per token. Open weights followed on July 26, with day-zero hosted access through providers including Together AI and Modal.
K3 moves Moonshot from a 256K window to a full 1M, which puts an open-weight model at parity with the proprietary frontier on context size.
Primary Use Cases
- Large codebase comprehension and multi-file refactoring
- Frontend development from visual designs and screenshots
- Self-hosted agentic workflows where weights need to stay in your infrastructure
Input Context Window – 1 million tokens
DeepSeek V4 reached general availability on July 20, 2026 after a preview period that began in April, and it makes a 1M context window the default floor across the entire lineup rather than a premium tier. V3.2 topped out at 128K.
V4-Pro runs $0.435/$0.87 per million input/output tokens and V4-Flash runs $0.14/$0.28, which is the cheapest 1M-token context available from any major provider by a wide margin. One caveat for production planning: API calls placed during Beijing peak hours carry a 2x price multiplier on the base rate.
Primary Use Cases
- Extended document summarization and deep Q&A over long texts
- Multi-step mathematical problem solving and chain-of-thought reasoning
- Cost-sensitive long-context workloads where off-peak batching is practical
Input Context Window – 983,616 tokens
Qwen 3.8 is Alibaba's 2.4 trillion parameter sparse MoE model, supporting just under 1M input tokens with a 131,072 token output window.
One thing to verify before you build on it: Alibaba has signalled that Qwen 3.8 will go open-weight, but no license file has been published yet, and both Qwen 3.7-Max and 3.6-Max-Preview shipped closed. Earlier open Qwen 3 releases used Apache 2.0. Treat the licensing as unsettled until Alibaba confirms it.
Primary Use Cases
- Complex reasoning, coding, and instruction-following tasks
- Multilingual and multi-domain production workflows
- Cost-efficient deployments where near-1M context is required
Input Context Window – Up to 1 million tokens
A 400B parameter MoE model with 17B active parameters and 128 experts, Maverick delivers flagship-level performance for enterprise applications while maintaining cost efficiency.
Primary Use Cases
- Enterprise-grade multimodal applications
- Advanced image and text understanding across 12 languages
- High-performance chat and assistant applications
Input Context Window – Up to 256,000 tokens
Mistral Large 3 is a 675B total / 41B active parameter sparse MoE model under Apache 2.0.
It delivers state-of-the-art performance with flexible enterprise deployment—on-premises, hybrid, and in-VPC—and is a top choice for teams with EU data residency requirements.
Primary Use Cases
- Professional coding and STEM workflows requiring high accuracy
- On-premise and hybrid deployments for regulated industries
- Multimodal understanding in enterprise settings with custom fine-tuning
What business cases are large context windows best for?
1. Coding and codebase analysis: Query, refactor, and audit entire code repositories faster.
Large context windows let developers load entire multi-repo codebases, documentation, and test files simultaneously—no chunking, no lost cross-file dependencies.
This is where the jump from 200K to 1M tokens has the most immediate practical impact. For a deeper look at which models developers use most for coding, see our roundup of the best LLMs for coding.
Examples
- Code completion and generation with full repository context, preserving style and architecture
- Refactoring and documentation across entire software systems in one session
- Security audits by reviewing an entire codebase for vulnerabilities at once
- Legacy code modernization: loading an entire COBOL system into context so Claude Code can map dependencies, flag risks, and generate modern equivalents without losing the thread across thousands of files
2. Agentic AI workflows: Run long-horizon agents that hold full task history in context.
Agentic systems make dozens or hundreds of tool calls per task—searching databases, reading files, executing code, verifying outputs. With a 1M context window, the entire trace stays intact: every tool call, observation, and intermediate reasoning step.
That eliminates the compaction and context-clearing that used to cause agents to lose the plot mid-task.
Examples
- Autonomous coding agents that plan, execute, and verify complex multi-file tasks without restarting
- Research agents synthesizing hundreds of papers, datasets, and codebases in a single pass
- AI project managers tracking decisions, dependencies, and progress across long engagements
3. Comprehensive document analysis: Analyze entire books, contracts, or research corpora without splitting.
Models retain full context throughout, making summarization, Q&A, and insight extraction more accurate than sliding-window approaches. Legal, financial, and research workflows benefit most.
Examples
- Financial reports analyzed end-to-end for trends and anomalies
- Legal contracts reviewed for risks, clauses, and inconsistencies across hundreds of pages
- Research literature synthesized across dozens of papers in a single request
4. Multimodal data processing: Handle large datasets combining text, images, and video.
Models like Gemini 3.1 Pro and Claude Opus 4.6 process text, images, video, and audio natively within the same context window—making them practical for multimedia workflows that previously required multiple separate pipelines.
Examples
- Medical imaging analyzed alongside patient history and clinical notes
- Video transcript analysis cross-referenced with supporting documentation for tagging or summarization
- Visual coding generating front-end interfaces directly from design screenshots or screen recordings
5. Enterprise knowledge management: Build systems that retrieve and reason across large internal document sets.
Large context windows reduce dependence on RAG pipelines for many use cases—entire knowledge bases, policy libraries, or case archives can be loaded directly, with the model reasoning across all of it at once.
Examples
- Corporate knowledge assistants referencing multiple internal documents to answer complex questions
- Legal research tools scanning thousands of cases for relevant precedents
- Healthcare AI consulting vast medical literature to support diagnosis or treatment decisions
Do long context windows cost more?
In short, yes. Long context windows enable advanced capabilities but do come at a higher computational and financial cost due to increased memory, slower processing, and more resource-heavy inference.
But, they don’t have to mean wasted money. When you have the right use case figured out and optimize your LLMs in production, you can control costs.
Here are the cost challenges you need to consider:
- Increased memory usage
Longer sequences mean more memory consumption. As the number of tokens in the context window increases, the model must store and process more information, which results in greater memory requirements.
This leads to higher GPU/TPU memory usage during inference and training. - Slower processing times
Processing longer inputs takes more time. Large context windows require the model to attend to more tokens, increasing the computational complexity.
Transformer models, like those used in GPT and similar architectures, use an attention mechanism where the complexity grows quadratically with the number of tokens. As the number of tokens increases, it significantly slows down the processing speed. - More expensive inference
Inference costs scale with input length. Models with larger context windows require more operations per token to maintain context over long inputs, resulting in higher compute costs for running predictions or generating outputs.
Cloud services, like OpenAI or Anthropic, usually charge based on the number of tokens processed, so longer contexts increase costs directly. - Higher energy and resource usage
More compute resources are needed for extended context handling. Handling longer sequences requires more powerful hardware to avoid bottlenecks.
Training and inference over large contexts might require higher-end GPUs, leading to higher operational costs, especially in large-scale deployments. - Optimization challenges
Models with larger context windows require more sophisticated optimization. Managing long sequences without performance degradation is a challenge.
Techniques like LongRoPE and other position encoding methods are used to improve efficiency, but these often come at an extra computational cost.
How to manage costs for large context LLMs?
- Use adaptive context windows: Instead of always using the maximum context window, some systems adapt the window size to the input length, reducing costs when smaller contexts suffice.
- Pruning or focusing attention: Techniques like sparse attention can help reduce the computational load by limiting attention to the most relevant tokens.
- Batching inputs: Combining shorter inputs in batches can help minimize resource use when long context windows aren’t required.
How do I hire senior AI engineers to build with large context window LLMs?
You could spend the next 6-18 months planning to recruit and build an AI team, but you won’t be building any AI capabilities. That’s why Codingscape exists.
We can assemble a senior AI development team for you in 4-6 weeks and start building your AI apps with large context LLMs. It’ll be faster to get started, more cost-efficient than internal hiring, and we’ll deliver high-quality results quickly.
Zappos, Twilio, and Veho are just a few companies that trust us to build their software and systems with a remote-first approach.
You can schedule a time to talk with us here. No hassle, no expectations, just answers.
Don't Miss
Another Update
new content is published
Cole
Cole is Codingscape's Content Marketing Strategist & Copywriter.