NEWHow AI models draw the world map · Gecko Tests
Beta
Context · 200K+

Cheapest 200K context LLMs

Every LLM with a 200,000+ token context window. Ranked by input price per 1M tokens.

Models40
Cheapest$0.00
Min context200K tokens
All pricingPricing home
What this page is
200K context is the modern default for frontier models. This page lists every priced 200K+ model, cheapest first. For long docs that fit in 200K, this tier offers the best balance of price, recall, and model diversity.

200K+ context models, cheapest first.

#ModelIn $/1MOut $/1MType
1DeepSeek logoDeepSeek V4 Flash 0731 (free)$0.00$0.00Closed
2openrouter logoElephant$0.00$0.00Closed
3openrouter logoFree Models Router$0.00$0.00Closed
4Google DeepMind logoGemma 4 26B A4B (free)$0.00$0.00OSS
5Google DeepMind logoGemma 4 31B (free)$0.00$0.00OSS
6tencent logoHy3 (free)$0.00$0.00Closed
7tencent logoHy3 preview (free)$0.00$0.00Closed
8Laguna M.1 (free)$0.00$0.00OSS
9Laguna XS.2 (free)$0.00$0.00OSS
10Google DeepMind logoLyria 3 Clip Preview$0.00$0.00Closed
11Google DeepMind logoLyria 3 Pro Preview$0.00$0.00Closed
12minimax logoMiniMax M3 (free)$0.00$0.00Closed
13NVIDIA logoNemotron 3 Nano 30B A3B (free)$0.00$0.00OSS
14NVIDIA logoNemotron 3 Nano Omni (free)$0.00$0.00OSS
15NVIDIA logoNemotron 3 Super (free)$0.00$0.00OSS
16NVIDIA logoNemotron 3 Ultra (free)$0.00$0.00OSS
17NVIDIA logoNemotron 3.5 Lightning (free)$0.00$0.00Closed
18nex-agi logoNex-N2-Pro (free)$0.00$0.00OSS
19nex-agi logoNex-N2.5-Mini (free)$0.00$0.00Closed
20nex-agi logoNex-N2.5-Pro (free)$0.00$0.00Closed
21Cohere logoNorth Mini Code (free)$0.00$0.00OSS
22openrouter logoOwl Alpha$0.00$0.00Closed
23Alibaba Qwen logoQwen3 Coder 480B A35B (free)$0.00$0.00OSS
24Alibaba Qwen logoQwen3 Next 80B A3B Instruct (free)$0.00$0.00OSS
25Alibaba Qwen logoQwen3.6 Plus (free)$0.00$0.00Closed
26Alibaba Qwen logoQwen3.6 Plus Preview (free)$0.00$0.00OSS
27Alibaba Qwen logoQwen3.8 27B (free)$0.00$0.00Closed
28stepfun logoStep 3.5 Flash (free)$0.00$0.00OSS
29Ling-2.6-flash$0.01$0.03Closed
30DeepSeek logoDeepSeek V4 Flash 0731$0.02$1.28Closed
31DeepSeek logoDeepSeek V4 Flash$0.03$1.28OSS
32OpenAI logoGPT-5 Nano (batch)$0.03$0.20Closed
33nex-agi logoNex-N2-Mini$0.03$0.10Closed
34nex-agi logoNex-N2.5-Mini$0.03$0.10Closed
35Alibaba Qwen logoQwen3.7 Flash$0.03$0.13Closed
36inception logoMercury 2.5$0.04$0.15Closed
37inception logoMercury 2.5 Preview$0.04$0.15Closed
38Alibaba Qwen logoQwen3 30B A3B Instruct 2507$0.05$0.19OSS
39OpenAI logoGPT-6 Luna (batch)$0.05$0.25Closed
40OpenAI logoGPT-6 Luna Pro (batch)$0.05$0.25Closed
Cheapest 200K
DeepSeek logo
DeepSeek V4 Flash 0731 (free)
input
$0.00/M
output
$0.00/M

DeepSeek V4 Flash 0731 (free) delivers 1.0M context at $0.00/M input. Great sweet spot for long docs without paying 1M context premium.

Runner up · 200K
openrouter logo
Elephant
input
$0.00/M
output
$0.00/M

Elephant delivers 262K context at $0.00/M input. Great sweet spot for long docs without paying 1M context premium.

Third · 200K
openrouter logo
Free Models Router
input
$0.00/M
output
$0.00/M

Free Models Router delivers 200K context at $0.00/M input. Great sweet spot for long docs without paying 1M context premium.

Cheapest
DeepSeek V4 Flash 0731 (free)
$0.00/M
$ per 1M input tokens
Why the gap

At the 200K tier, premium pricing buys better tail-end retrieval and higher overall reasoning. For retrieval-heavy RAG, cheap models often match premium ones.

Most expensive
GPT-6 Luna (batch)
$0.05/M
$ per 1M input tokens
Yes. 200K handles a full novel, a medium-sized codebase, or 100+ pages of PDFs. Only use 1M when you truly need whole-repo or multi-book context.