Skip to content
LMC Feed-Models, Papers, Benchmarks. Zero Fluff.Live

Context Window Explorer

Compare 450 AI models by context window size. Find the models that can handle the longest inputs, from 10M+ token contexts down to compact windows under 32K.

Total Models

450

Largest Context

2M

Average Context

490K

128K+ Models

384

Context Window Size Distribution

LMMarketCap.com

Tier Distribution

How 450 models are distributed across context window tiers.

10M+

0

0% of models

1M+

157

35% of models

256K+

114

25% of models

128K+

113

25% of models

32K+

27

6% of models

Under 32K

39

9% of models

All Models by Context Window

450 models sorted by context window size, largest first.

ModelContextScore
Grok 4.20 Multi-AgentxAI2M88
Grok 4.20xAI2M89
GLM Flash Latest~z-ai1.3M40
GLM 5.3 FlashZhipu AI1.3M78
GLM Latest~z-ai1.3M40
GLM 5.3Zhipu AI1.3M79
DeepSeek V4 Flash Latest~deepseek1.3M40
DeepSeek V4 Flash 0731DeepSeek1.3M40
Llama 4 ScoutMeta1.3M60
GPT Astra Latest~openai1.1M40
GPT Sol Latest~openai1.1M40
GPT Terra Latest~openai1.1M40
GPT Luna Latest~openai1.1M40
GPT-6 AstraOpenAI1.1M82
GPT-6 Astra (batch)OpenAI1.1M82
GPT-6 Astra ProOpenAI1.1M82
GPT-6 Astra Pro (batch)OpenAI1.1M82
GPT-5.6 Luna ProOpenAI1.1M89
GPT-5.6 Luna Pro (batch)OpenAI1.1M89
GPT-5.6 LunaOpenAI1.1M89
GPT-5.6 Luna (batch)OpenAI1.1M89
GPT-5.6 Terra ProOpenAI1.1M89
GPT-5.6 Terra Pro (batch)OpenAI1.1M89
GPT-5.6 TerraOpenAI1.1M89
GPT-5.6 Terra (batch)OpenAI1.1M89
GPT-5.6 Sol ProOpenAI1.1M89
GPT-5.6 Sol Pro (batch)OpenAI1.1M89
GPT-5.6 SolOpenAI1.1M89
GPT-5.6 Sol (batch)OpenAI1.1M89
GPT-5.5 ProOpenAI1.1M93
GPT-5.5 Pro (batch)OpenAI1.1M93
GPT-5.5OpenAI1.1M93
GPT-5.5 (batch)OpenAI1.1M93
MiMo-V2.5-ProXiaomi1.1M76
MiMo-V2.5Xiaomi1.1M73
GPT-5.4 ProOpenAI1.1M92
GPT-5.4 Pro (batch)OpenAI1.1M92
GPT-5.4OpenAI1.1M92
GPT-5.4 (batch)OpenAI1.1M92
LongCat 2.0Meituan1.0M40
MiMo-V2.6-Pro-UltraSpeedXiaomi1.0M40
MiMo-V2.6-FlashXiaomi1.0M40
MiMo-V2.6-ProXiaomi1.0M76
GLM 5.3 FlashXZhipu AI1.0M78
DeepSeek Pro Latest~deepseek1.0M40
DeepSeek Flash Latest~deepseek1.0M40
DeepSeek V4.1 FlashDeepSeek1.0M40
Muse Spark 1.3 Contributormeta1.0M40
Muse Spark 1.3meta1.0M81
Gemini 3.8 FlashGoogle1.0M81
Gemini 3.8 Flash (batch)Google1.0M81
Hy4 previewTencent1.0M40
GLM 5.3 Flash (batch)Zhipu AI1.0M78
Muse Spark 1.2 Contributormeta1.0M40
DeepSeek V4 Flash Vision ExpDeepSeek1.0M40
DeepSeek V4 Flash Vision Exp (batch)DeepSeek1.0M40
GLM 5.3 (batch)Zhipu AI1.0M79
Gemini 3.7 FlashGoogle1.0M40
Gemini 3.7 Flash (batch)Google1.0M40
Qwen3.8 2.4T A95BAlibaba1.0M40
DeepSeek V4 Pro 0813DeepSeek1.0M40
DeepSeek V4 Pro 0813 (batch)DeepSeek1.0M40
Muse Spark 1.2meta1.0M81
DeepSeek V4 Flash 0731 (batch)DeepSeek1.0M40
Inkling Smallthinkingmachines1.0M69
Inkling Small (free)thinkingmachines1.0M69
Laguna S 2.1poolside1.0M40
Gemini 3.6 FlashGoogle1.0M40
Gemini 3.6 Flash (batch)Google1.0M40
Gemini 3.5 Flash LiteGoogle1.0M77
Gemini 3.5 Flash Lite (batch)Google1.0M77
Inklingthinkingmachines1.0M73
Inkling (free)thinkingmachines1.0M73
Kimi K3Moonshot AI1.0M40
Muse Spark 1.1meta1.0M81
GLM 5.2Zhipu AI1.0M79
GLM 5.2 (batch)Zhipu AI1.0M79
MiniMax M3MiniMax1.0M74
Gemini 3.5 FlashGoogle1.0M79
Gemini 3.5 Flash (batch)Google1.0M79
Gemini 3.1 Flash LiteGoogle1.0M--
Gemini 3.1 Flash Lite (batch)Google1.0M--
Gemini Pro Latest~google1.0M--
Kimi Latest~moonshotai1.0M--
Gemini Flash Latest~google1.0M--
DeepSeek V4 Pro 0423DeepSeek1.0M--
DeepSeek V4 Flash 0423DeepSeek1.0M--
Lyria 3 Pro PreviewGoogle1.0M--
Lyria 3 Clip PreviewGoogle1.0M--
Gemini 3.1 Flash Lite PreviewGoogle1.0M79
Gemini 3.1 Pro Preview Custom ToolsGoogle1.0M92
Gemini 3.1 Pro PreviewGoogle1.0M92
Gemini 3.1 Pro Preview (batch)Google1.0M92
Gemini 3 Flash PreviewGoogle1.0M88
Gemini 3 Flash Preview (batch)Google1.0M88
Gemini 2.5 Flash LiteGoogle1.0M79
Gemini 2.5 Flash Lite (batch)Google1.0M79
Gemini 2.5 FlashGoogle1.0M79
Gemini 2.5 Flash (batch)Google1.0M79
Gemini 2.5 ProGoogle1.0M84
Gemini 2.5 Pro (batch)Google1.0M84
Gemini 2.5 Pro Preview 06-05Google1.0M84
Llama 4 MaverickMeta1.0M71
GPT-4.1OpenAI1.0M68
GPT-4.1 (batch)OpenAI1.0M68
GPT-4.1 MiniOpenAI1.0M56
GPT-4.1 Mini (batch)OpenAI1.0M59
GPT-4.1 NanoOpenAI1.0M42
GPT-4.1 Nano (batch)OpenAI1.0M59
Palmyra X5Writer1.0M--
MiniMax-01MiniMax1.0M--
Fugu Ultra v2sakana1M40
Fugu Maxsakana1M40
Qwen3.8 Max (0902)Alibaba1M76
Claude Fable 5.1Anthropic1M96
Claude Fable 5.1 (batch)Anthropic1M96
Qwen3.8 FlashAlibaba1M40
Qwen3.8 27BAlibaba1M73
Nemotron 3.5 Lightning (free)NVIDIA1M40
Qwen3.7 FlashAlibaba1M40
Claude Opus 5Anthropic1M95
Claude Opus 5 (batch)Anthropic1M40
Claude Sonnet 5Anthropic1M85
Claude Sonnet 5 (batch)Anthropic1M40
Fugu Ultrasakana1M40
Claude Fable Latest~anthropic1M--
Claude Fable 5Anthropic1M96
Claude Fable 5 (batch)Anthropic1M96
Nemotron 3 Ultra (free)NVIDIA1M--
Qwen3.7 PlusAlibaba1M76
Claude Opus 4.8Anthropic1M95
Claude Opus 4.8 (batch)Anthropic1M95
Qwen3.7 MaxAlibaba1M75
Grok 4.3xAI1M89
Grok 4.3 (batch)xAI1M81
Claude Sonnet Latest~anthropic1M--
Qwen3.5 Plus 2026-04-20Alibaba1M75
Qwen3.6 FlashAlibaba1M--
Claude Opus Latest~anthropic1M--
Claude Opus 4.7Anthropic1M95
Claude Opus 4.7 (batch)Anthropic1M95
Qwen3.6 PlusAlibaba1M74
Qwen3.5-FlashAlibaba1M69
Claude Sonnet 4.6Anthropic1M85
Claude Sonnet 4.6 (batch)Anthropic1M85
Qwen3.5 Plus 2026-02-15Alibaba1M68
Claude Opus 4.6Anthropic1M90
Claude Opus 4.6 (batch)Anthropic1M90
Nova 2 LiteAmazon1M60
Nova Premier 1.0Amazon1M--
Claude Sonnet 4.5Anthropic1M82
Claude Sonnet 4.5 (batch)Anthropic1M82
Qwen3 Coder PlusAlibaba1M--
Qwen3 Coder FlashAlibaba1M--
Qwen Plus 0728Alibaba1M--
MiniMax M1MiniMax1M71
Qwen-PlusAlibaba1M--
Solar Pro 4Upstage524K40
Dots3-Note Preview (free)dots-studio512K40
Grok 4.7xAI500K89
Grok 4.6xAI500K89
Grok 4.5xAI500K89
Grok Latest~x-ai500K40
GPT Chat LatestOpenAI400K--
GPT Mini Latest~openai400K--
GPT-5.4 NanoOpenAI400K79
GPT-5.4 Nano (batch)OpenAI400K79
GPT-5.4 MiniOpenAI400K79
GPT-5.4 Mini (batch)OpenAI400K79
GPT-5.3-CodexOpenAI400K91
GPT-5.2-CodexOpenAI400K91
GPT-5.2 ProOpenAI400K91
GPT-5.2 Pro (batch)OpenAI400K91
GPT-5.2OpenAI400K91
GPT-5.2 (batch)OpenAI400K91
GPT-5.1-Codex-MaxOpenAI400K89
GPT-5.1OpenAI400K89
GPT-5.1 (batch)OpenAI400K88
GPT-5.1-CodexOpenAI400K89
GPT-5.1-Codex-MiniOpenAI400K88
GPT-5 Image MiniOpenAI400K--
GPT-5 ImageOpenAI400K--
GPT-5 ProOpenAI400K89
GPT-5 Pro (batch)OpenAI400K89
GPT-5OpenAI400K89
GPT-5 (batch)OpenAI400K89
GPT-5 MiniOpenAI400K54
GPT-5 Mini (batch)OpenAI400K77
GPT-5 NanoOpenAI400K62
GPT-5 Nano (batch)OpenAI400K77
Nova Lite 1.0Amazon300K--
Nova Pro 1.0Amazon300K--
GPT-5.4 Image 2OpenAI272K--
Ternary Bonsai 2 27Bprism-ml262K40
Paretounbiased262K40
Ling 3.0 Flash VL (free)inclusionai262K40
Ling 3.0 Flash Sante (free)inclusionai262K40
Ling 3.0 Flash Fininclusionai262K40
Ling 3.0 Flash Fin (free)inclusionai262K40
Qwen3.8 27B (free)Alibaba262K73
Seed 2.1 TurboByteDance262K40
Seed-2.0-CodeByteDance262K40
Nemotron 3.5 LightningNVIDIA262K40
Sakana Namazusakana262K40
Ling 3.0 Flashinclusionai262K40
Laguna S 2.1 (free)poolside262K40
KAT-Coder-Pro V2.5Kuaishou262K40
Hy3Tencent262K74
Laguna XS 2.1poolside262K40
Laguna XS 2.1 (free)poolside262K40
Kimi K2.7 CodeMoonshot AI262K40
Nemotron 3 UltraNVIDIA262K--
Step 3.7 FlashStepFun262K--
Mistral Medium 3.5Mistral AI262K72
Mistral Medium 3.5 (batch)Mistral AI262K--
Qwen3.6 35B A3BAlibaba262K--
Qwen3.6 Max PreviewAlibaba262K75
Qwen3.6 27BAlibaba262K--
Hy3 previewTencent262K68
Kimi K2.6Moonshot AI262K76
Gemma 4 26B A4B Google262K73
Gemma 4 26B A4B (free)Google262K73
Gemma 4 31BGoogle262K81
Gemma 4 31B (free)Google262K81
Trinity Large Thinkingarcee-ai262K63
Mistral Small 4Mistral AI262K--
Mistral Small 4 (batch)Mistral AI262K--
Nemotron 3 SuperNVIDIA262K--
Nemotron 3 Super (free)NVIDIA262K--
Seed-2.0-LiteByteDance262K--
Qwen3.5-9BAlibaba262K67
Seed-2.0-MiniByteDance262K--
Qwen3.5-35B-A3BAlibaba262K76
Qwen3.5-27BAlibaba262K77
Qwen3.5-122B-A10BAlibaba262K78
Qwen3.5 397B A17BAlibaba262K79
Qwen3 Max ThinkingAlibaba262K68
Qwen3 Coder NextAlibaba262K--
Step 3.5 FlashStepFun262K66
Kimi K2.5Moonshot AI262K48
Seed 1.6 FlashByteDance262K--
Seed 1.6ByteDance262K--
Nemotron 3 Nano 30B A3BNVIDIA262K--
Devstral 2 2512Mistral AI262K--
Ministral 3 14B 2512Mistral AI262K--
Ministral 3 8B 2512Mistral AI262K--
Ministral 3 8B 2512 (batch)Mistral AI262K--
Mistral Large 3 2512 (batch)Mistral AI262K67
Kimi K2 ThinkingMoonshot AI262K--
Qwen3 VL 8B InstructAlibaba262K--
Qwen3 VL 30B A3B ThinkingAlibaba262K--
Qwen3 VL 30B A3B InstructAlibaba262K--
Qwen3 VL 235B A22B InstructAlibaba262K69
Qwen3 MaxAlibaba262K67
Qwen3 Next 80B A3B ThinkingAlibaba262K67
Qwen3 Next 80B A3B InstructAlibaba262K67
Kimi K2 0905Moonshot AI262K--
Qwen3 Coder 30B A3B InstructAlibaba262K--
Qwen3 30B A3B Instruct 2507Alibaba262K64
Qwen3 Coder 480B A35BAlibaba262K--
Qwen3 235B A22B Instruct 2507Alibaba262K65
Falcon-H1-Arabic 34B InstructTII262K--
Falcon-H1-Arabic 7B InstructTII262K--
Mercury 2.5Inception260K61
North Mini Code (free)Cohere256K40
Grok Build 0.1xAI256K--
Nemotron 3 Nano Omni (free)NVIDIA256K--
Codestral 2508Mistral AI256K--
Codestral 2508 (batch)Mistral AI256K--
Mistral Small 3.2 24BMistral AI256K--
Command ACohere256K51
GLM 5.1Zhipu AI205K78
MiniMax M2.7MiniMax205K72
MiniMax M2.5MiniMax205K72
GLM 5Zhipu AI205K78
MiniMax M2.1MiniMax205K71
GLM 4.7Zhipu AI205K75
MiniMax M2MiniMax205K71
GLM 4.6Zhipu AI205K75
GLM 5V TurboZhipu AI203K72
GLM 5 TurboZhipu AI203K78
Claude Haiku Latest~anthropic200K--
GLM 4.7 FlashZhipu AI200K64
Claude Opus 4.5Anthropic200K85
Claude Opus 4.5 (batch)Anthropic200K85
Sonar Pro SearchPerplexity200K--
Claude Haiku 4.5Anthropic200K70
Claude Haiku 4.5 (batch)Anthropic200K70
Claude Opus 4.1Anthropic200K74
Claude Opus 4.1 (batch)Anthropic200K75
o3 ProOpenAI200K87
Claude Sonnet 4Anthropic200K74
o4 Mini HighOpenAI200K72
o3OpenAI200K87
o3 (batch)OpenAI200K87
o4 MiniOpenAI200K81
o4 Mini (batch)OpenAI200K81
o1-proOpenAI200K74
Sonar ProPerplexity200K--
o3 Mini HighOpenAI200K64
o3 MiniOpenAI200K75
o3 Mini (batch)OpenAI200K74
o1OpenAI200K74
Claude 3 HaikuAnthropic200K51
Composer 2Cursor200K66
Composer 2 FastCursor200K66
DeepSeek V3.2DeepSeek164K83
DeepSeek V3.2 ExpDeepSeek164K72
DeepSeek V3.1 TerminusDeepSeek164K69
DeepSeek V3.1DeepSeek164K72
R1 0528DeepSeek164K79
Llama Guard 4 12BMeta164K--
DeepSeek V3 0324DeepSeek164K72
DeepSeek V3DeepSeek164K70
Ling 3.0 Flash VLinclusionai131K40
Granite 4.2 8BIBM131K54
Muse Glimmer 30Bmeta131K40
Muse Glimmer 30B (batch)meta131K40
Aion-3.0-Miniaion-labs131K40
Aion-3.0aion-labs131K40
Nano Banana 2 (Gemini 3.1 Flash Image)Google131K--
Nano Banana Pro (Gemini 3 Pro Image)Google131K--
Nemotron 3.5 Content SafetyNVIDIA131K--
Aion-2.0aion-labs131K--
Solar Pro 3Upstage131K--
GLM 4.6VZhipu AI131K66
Ministral 3 3B 2512Mistral AI131K--
gpt-oss-safeguard-20bOpenAI131K--
Qwen3 VL 32B InstructAlibaba131K--
Qwen3 VL 8B ThinkingAlibaba131K--
Qwen3 VL 235B A22B ThinkingAlibaba131K69
Mistral Medium 3.1Mistral AI131K--
Mistral Medium 3.1 (batch)Mistral AI131K--
gpt-oss-120bOpenAI131K62
gpt-oss-20bOpenAI131K57
GLM 4.5Zhipu AI131K75
GLM 4.5 AirZhipu AI131K71
Qwen3 235B A22B Thinking 2507Alibaba131K65
Kimi K2 0711Moonshot AI131K--
Hunyuan A13B InstructTencent131K--
Mistral Medium 3Mistral AI131K--
Qwen3 30B A3BAlibaba131K64
Qwen3 8BAlibaba131K61
Qwen3 14BAlibaba131K--
Qwen3 32BAlibaba131K--
Qwen3 235B A22BAlibaba131K54
Gemma 3 4BGoogle131K--
Gemma 3 12BGoogle131K--
Gemma 3 27BGoogle131K--
Llama 3.3 70B InstructMeta131K67
Mistral Large 2407Mistral AI131K66
Llama 3.2 3B InstructMeta131K--
Llama 3.1 70B InstructMeta131K65
Llama 3.1 8B InstructMeta131K45
Mistral NemoMistral AI131K--
Falcon-H1-Arabic 3B InstructTII131K--
Granite 4.0 MicroIBM131K--
Schematron V2 Turboinference-net128K40
Schematron V2 Smallinference-net128K40
Nemotron 3.5 Content Safety (free)NVIDIA128K--
Mercury 2Inception128K61
GPT AudioOpenAI128K--
GPT Audio MiniOpenAI128K--
GPT-5.2 ChatOpenAI128K91
UI-TARS 7B ByteDance128K--
Mistral Small 3.1 24BMistral AI128K--
Sonar Reasoning ProPerplexity128K--
Sonar Deep ResearchPerplexity128K--
Qwen2.5 VL 72B InstructAlibaba128K--
Command R7B (12-2024)Cohere128K--
Nova Micro 1.0Amazon128K--
GPT-4o (2024-11-20)OpenAI128K71
Command R (08-2024)Cohere128K49
Command R+ (08-2024)Cohere128K49
GPT-4o (2024-08-06)OpenAI128K71
GPT-4o-miniOpenAI128K69
GPT-4o-mini (2024-07-18)OpenAI128K57
GPT-4o-mini (batch)OpenAI128K63
GPT-4oOpenAI128K71
GPT-4o (2024-05-13)OpenAI128K71
GPT-4o (batch)OpenAI128K72
GPT-4 TurboOpenAI128K67
GPT-4 Turbo (batch)OpenAI128K62
Mistral LargeMistral AI128K66
SonarPerplexity127K--
ERNIE 4.5 VL 424B A47B Baidu123K--
Qwen3 30B A3B Thinking 2507Alibaba82K64
LFM2.5-2.6B (free)Liquid AI66K40
Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)Google66K--
Nano Banana 2 (Gemini 3.1 Flash Image Preview)Google66K--
MiniMax M2-herMiniMax66K68
Nano Banana Pro (Gemini 3 Pro Image Preview)Google66K--
GLM 4.5VZhipu AI66K62
Reka Flash 3rekaai66K--
Mixtral 8x22B InstructMistral AI66K63
WizardLM-2 8x22BMicrosoft66K--
R1DeepSeek64K74
Llama 3.2 1B InstructMeta60K--
GLM 5.2 (free)Zhipu AI33K76
Perceptron Mk1perceptron33K--
Voxtral Small 24B 2507Mistral AI33K--
Nano Banana (Gemini 2.5 Flash Image)Google33K--
SabaMistral AI33K--
Mistral Small 3Mistral AI33K--
Qwen2.5 Coder 32B InstructAlibaba33K--
Qwen2.5 7B InstructAlibaba33K--
Qwen2.5 72B InstructAlibaba33K--
Falcon Arabic 7B InstructTII33K--
Falcon3 10B InstructTII33K--
Falcon3 7B InstructTII33K--
Falcon Mamba 7B InstructTII33K--
GPT-3.5 Turbo 16kOpenAI16K--
GPT-3.5 TurboOpenAI16K--
GPT-3.5 Turbo (batch)OpenAI16K--
Reka Edgerekaai16K--
Phi 4Microsoft16K60
Hy-MT2-1.8BTencent8K40
Hy-MT2-30B-A3BTencent8K40
Hy-MT2-7BTencent8K40
R1 Distill Llama 70BDeepSeek8K41
Gemma 2 27BGoogle8K77
GPT-4OpenAI8K65
ALLaM 7B Instruct (preview)HUMAIN4K--
ALLaM 1 13B InstructHUMAIN4K--
ALLaM 2 7B InstructHUMAIN4K--
ALLaM 34BHUMAIN4K--
GPT-3.5 Turbo (older v0613)OpenAI4K--
GPT-3.5 Turbo InstructOpenAI4K--
Midjourney v6.1Midjourney0--
DALL-E 3OpenAI0--
Stable Diffusion 3.5Stability AI0--
FLUX.1 ProBlack Forest Labs0--
Ideogram 2.0Ideogram0--
Recraft V3Recraft0--
Imagen 3Google0--
Adobe Firefly 3Adobe0--
Leonardo PhoenixLeonardo AI0--
SoraOpenAI0--
Veo 2Google0--
Runway Gen-3 AlphaRunway0--
Kling 1.6Kuaishou0--
Pika 2.0Pika0--
MiniMax Video-01MiniMax0--
Luma Dream MachineLuma AI0--
Stable Video DiffusionStability AI0--
Wan 2.1 T2VWan AI0--
LTX-Video 2Lightricks0--
SWE-1.5Windsurf0--
autofixer-01Vercel0--
MellumJetBrains0--

Use Cases by Context Size

Different context window sizes unlock different capabilities. Here is what each tier is best suited for.

10M+

Entire Codebases & Libraries

0 models
  • •Analyze a complete monorepo with hundreds of files
  • •Process entire books or multi-volume document sets
  • •Ingest full API documentation suites at once
1M+

Large-Scale Analysis

157 models
  • •Review and refactor full Git repositories
  • •Process lengthy legal contracts and regulatory documents
  • •Multi-document research synthesis across dozens of papers
256K+

Comprehensive Document Work

114 models
  • •Analyze long technical specifications
  • •Process extensive conversation histories with full context
  • •Review multiple related code files simultaneously
128K+

Standard Long-Context Tasks

113 models
  • •Code review across multiple files
  • •Summarize research papers and reports
  • •Extended multi-turn conversations with memory
32K+

General Purpose

27 models
  • •Standard coding assistance and debugging
  • •Document editing and content creation
  • •Typical question-answering and chat interactions
Under 32K

Quick Tasks

39 models
  • •Simple completions and classifications
  • •Short Q&A and factual lookups
  • •Lightweight text transformations

Explore More

Compare capabilities, pricing, or dive into head-to-head model comparisons.

All ExplorersCapabilitiesPricingContext Compare ToolLeaderboard
Frequently Asked Questions

The context window is the maximum amount of text (measured in tokens) an AI model can process in a single conversation. Larger context windows allow processing longer documents, maintaining longer conversation history, and analyzing more data at once.

Gemini 2.0 Pro supports up to 2 million tokens. Claude 3.5 Sonnet offers 200K tokens. GPT-4o provides 128K tokens. Larger context windows generally cost more per request.

Yes - processing more tokens costs more since you pay per token. However, larger context windows enable use cases like full-codebase analysis and long document Q&A that are impossible with smaller windows.

Context Window Explorer - Compare AI Model Context Sizes | LM Market Cap