Skip to content
OpenKey

Long-context models

Models with a context window of 400K tokens or more. 87 of 338 catalog models match as of July 2026.

87
Models
$0.052
Input from /1M
10M
Max context
Jun 30, 2026
Newest release

87 models in OpenKey's catalog now clear the 1M-token-context bar, and the newest, Claude Sonnet 5, landed 2026-06-30 with a 1,000,000-token window at $2/$10 per million (input/output). Pricing spans an enormous range depending on how you filter out the pay-as-you-go router products, but real-model input costs run from $0.09/M (DeepSeek V4 Flash) up to $30/M (GPT-5.5 Pro, Claude Opus 4.7 Fast). Several models now push past 1M tokens entirely — Grok 4.20 and Grok 4.20 Multi-Agent sit at 2,000,000, and Pareto Code Router claims 2,000,000 as well. If you're building something that ingests full codebases or long documents, this list is where to start.

Top picks

DeepSeek V4 Flash

Cheapest long-context model with a verified benchmark result in this set: $0.09/M input, $0.18/M output, 1,048,576 tokens of context, and a design_arena SVG rank of 23 (elo 1216). If you need scale on a budget, this is the floor.

Claude Opus 4.7

1,000,000-token context at $5/M input, $25/M output, and the top design_arena agenticgamedev rank (rank 1, elo 1289) in this list — the strongest benchmark-to-context combination among the fee-based models.

Grok 4.20

2,000,000-token context, double most competitors, at $1.25/M input and $2.5/M output — the largest window here at a mid-tier price, useful when you genuinely need more than 1M tokens of room.

All 87 models

Full catalog →
  • Claude Sonnet 5ctx 1M · inout
  • Fugu Ultractx 1M · inout
  • GLM 5.2ctx 1.0M · inout
  • Fusionctx 1M · inVariableoutVariable
  • Claude Fable 5ctx 1M · inout
  • Nemotron 3 Ultractx 1M · inout
  • Nemotron 3 Ultra (free)Freectx 1M · inFreeoutFree
  • Qwen3.7 Plusctx 1M · inout
  • MiniMax M3ctx 1.0M · inout
  • Claude Opus 4.8ctx 1M · inout
  • Claude Opus 4.8 (Fast)ctx 1M · inout
  • Qwen3.7 Maxctx 1M · inout
  • Gemini 3.5 Flashctx 1.0M · inout
  • Claude Opus 4.7 (Fast)ctx 1M · inout
  • Gemini 3.1 Flash Litectx 1.0M · inout
  • GPT Chat Latestctx 400K · inout
  • Grok 4.3ctx 1M · inout
  • Qwen3.5 Plus 2026-04-20ctx 1M · inout
  • Qwen3.6 Flashctx 1M · inout
  • DeepSeek V4 Flashctx 1.0M · inout
  • DeepSeek V4 Proctx 1.0M · inout
  • GPT-5.5ctx 1.1M · inout
  • GPT-5.5 Proctx 1.1M · inout
  • MiMo-V2.5ctx 1.0M · inout
  • MiMo-V2.5-Proctx 1.0M · inout
  • Pareto Code Routerctx 2M · inVariableoutVariable
  • Claude Opus 4.7ctx 1M · inout
  • Qwen3.6 Plusctx 1M · inout
  • Grok 4.20ctx 2M · inout
  • Grok 4.20 Multi-Agentctx 2M · inout
  • Lyria 3 Clip PreviewFreectx 1.0M · inFreeoutFree
  • Lyria 3 Pro PreviewFreectx 1.0M · inFreeoutFree
  • GPT-5.4 Minictx 400K · inout
  • GPT-5.4 Nanoctx 400K · inout
  • Nemotron 3 Superctx 1M · inout
  • Nemotron 3 Super (free)Freectx 1M · inFreeoutFree
  • GPT-5.4ctx 1.1M · inout
  • GPT-5.4 Proctx 1.1M · inout
  • Gemini 3.1 Flash Lite Previewctx 1.0M · inout
  • Gemini 3.1 Pro Preview Custom Toolsctx 1.0M · inout
  • Qwen3.5-Flashctx 1M · inout
  • GPT-5.3-Codexctx 400K · inout
  • Gemini 3.1 Pro Previewctx 1.0M · inout
  • Claude Sonnet 4.6ctx 1M · inout
  • Qwen3.5 Plus 2026-02-15ctx 1M · inout
  • Claude Opus 4.6ctx 1M · inout
  • Palmyra X5ctx 1.0M · inout
  • GPT-5.2-Codexctx 400K · inout
  • Gemini 3 Flash Previewctx 1.0M · inout
  • GPT-5.2ctx 400K · inout
  • GPT-5.2 Proctx 400K · inout
  • GPT-5.1-Codex-Maxctx 400K · inout
  • Nova 2 Litectx 1M · inout
  • GPT-5.1ctx 400K · inout
  • GPT-5.1-Codexctx 400K · inout
  • GPT-5.1-Codex-Minictx 400K · inout
  • Nova Premier 1.0ctx 1M · inout
  • GPT-5 Image Minictx 400K · inout
  • GPT-5 Imagectx 400K · inout
  • GPT-5 Proctx 400K · inout
  • Claude Sonnet 4.5ctx 1M · inout
  • Gemini 2.5 Flash Lite Preview 09-2025ctx 1.0M · inout
  • GPT-5 Codexctx 400K · inout
  • Qwen3 Coder Plusctx 1M · inout
  • Qwen3 Coder Flashctx 1M · inout
  • Qwen Plus 0728ctx 1M · inout
  • Qwen Plus 0728 (thinking)ctx 1M · inout
  • GPT-5ctx 400K · inout
  • GPT-5 Minictx 400K · inout
  • GPT-5 Nanoctx 400K · inout
  • Qwen3 Coder 480B A35Bctx 1.0M · inout
  • Qwen3 Coder 480B A35B (free)Freectx 1.0M · inFreeoutFree
  • Gemini 2.5 Flash Litectx 1.0M · inout
  • Gemini 2.5 Flashctx 1.0M · inout
  • Gemini 2.5 Proctx 1.0M · inout
  • MiniMax M1ctx 1M · inout
  • Gemini 2.5 Pro Preview 06-05ctx 1.0M · inout
  • Claude Sonnet 4ctx 1M · inout
  • Gemini 2.5 Pro Preview 05-06ctx 1.0M · inout
  • GPT-4.1ctx 1.0M · inout
  • GPT-4.1 Minictx 1.0M · inout
  • GPT-4.1 Nanoctx 1.0M · inout
  • Llama 4 Maverickctx 1.0M · inout
  • Llama 4 Scoutctx 10M · inout
  • Qwen-Plusctx 1M · inout
  • MiniMax-01ctx 1.0M · inout
  • Auto Routerctx 2M · inVariableoutVariable

Prices per 1M tokens, flat 3% fee included. Hover or tap a price for the math.

How to choose

First tradeoff: context size versus real cost. Grok 4.20 and Grok 4.20 Multi-Agent double the window to 2,000,000 tokens at $1.25/$2.50 per M — cheap insurance if you're not sure how long your inputs will run. Second: modality matters more than the spec sheet suggests. Gemini 3.5 Flash and Gemini 3.1 Flash Lite handle text, image, file, audio, and video in one call; DeepSeek V4 Flash and Nemotron 3 Super are text-only but far cheaper per token. Third: don't pay Opus-tier prices ($30/M input on GPT-5.5 Pro, Claude Opus 4.7 Fast) unless the benchmark backs it up — Claude Opus 4.7 gets a rank-1 agenticgamedev result at a sixth of that input price.

Questions

What's the cheapest long-context model on OpenKey?
DeepSeek V4 Flash, at $0.09/M input and $0.18/M output for 1,048,576 tokens of context. On OpenKey that's $0.0927/M input and $0.1854/M output after the flat 3% fee ($0.09 × 1.03, $0.18 × 1.03). It also carries a real benchmark result (design_arena SVG, rank 23).
Which model has the largest context window in this list?
Grok 4.20 and Grok 4.20 Multi-Agent both sit at 2,000,000 tokens, tied with the Pareto Code Router product. Among priced, benchmarked chat models, that's double the 1,000,000-token window most competitors ship.
Are there free long-context models?
Yes. Nemotron 3 Ultra (free) and Nemotron 3 Super (free) both offer 1,000,000-token context at $0/M for input and output — the paid versions carry the same specs and benchmark scores, so the free tier isn't a stripped-down variant here.