Cheapest AI model for long-context (128K+)

Models with 128K-token or larger context windows, sorted cheapest-first.

Token Price Index › Browse › Long context

401 models matched for long context tasks. Cheapest right now: DeepSeek: DeepSeek V4.1 Flash at $0.024/M blended $/M. Sorted cheapest-first, updated daily.

Subscribe — $9/mo
CodingSummarizationChatVisionEmbeddingsReasoningCreative writingBrowse all
ModelContextInput $/MOutput $/MBlended $/M
DeepSeek: DeepSeek V4.1 Flashvisioncache
deepseek
1,048,576$0.028/M$0.011/M$0.024/M
Mistral: Mistral Nemo
mistralai
131,072$0.029/M$0.030/M$0.029/M
ScaleDown: Summarizevision
scaledown
1,000,000$0.040/MFree$0.030/M
ScaleDown: Extractvision
scaledown
1,000,000$0.040/MFree$0.030/M
inclusionAI: Ling 3.0 Flashcache
inclusionai
262,144$0.021/M$0.063/M$0.032/M
DeepSeek: DeepSeek V4 Flash Latestcache
~deepseek
1,048,576$0.0039/M$0.120/M$0.033/M
OpenAI: gpt-oss-20bcache
openai
131,072$0.018/M$0.090/M$0.036/M
IBM: Granite 4.0 Micro
ibm-granite
131,000$0.017/M$0.112/M$0.041/M
Nex AGI: Nex-N2.5-Minivisioncache
nex-agi
262,144$0.025/M$0.100/M$0.044/M
OpenAI: gpt-oss-20b (batch)
openai
131,072$0.024/M$0.112/M$0.046/M
Qwen: Qwen3.7 Flashvisioncache
qwen
1,000,000$0.030/M$0.130/M$0.055/M
OpenAI: gpt-oss-120b (batch)
openai
131,072$0.030/M$0.136/M$0.056/M
Meta: Llama 3.1 8B Instructcache
meta-llama
131,072$0.050/M$0.080/M$0.058/M
Inference.net: Schematron V2 Turbocache
inference-net
128,000$0.030/M$0.150/M$0.060/M
Amazon: Nova Micro 1.0
amazon
128,000$0.035/M$0.140/M$0.061/M
Google: Gemma 3 4Bvision
google
131,072$0.050/M$0.100/M$0.062/M
Cohere: Command R7B (12-2024)
cohere
128,000$0.037/M$0.150/M$0.066/M
Inception: Mercury 2.5cache
inception
260,000$0.040/M$0.150/M$0.068/M
OpenAI: GPT-5 Nano (batch)visioncache
openai
400,000$0.025/M$0.200/M$0.069/M
DeepSeek: DeepSeek V4.1 Flash (batch)visioncache
deepseek
1,048,576$0.050/M$0.150/M$0.075/M
Poolside: Laguna XS 2.1cache
poolside
262,144$0.060/M$0.120/M$0.075/M
Mistral: Ministral 3 8B 2512 (batch)visioncache
mistralai
262,144$0.075/M$0.075/M$0.075/M
Google: Gemma 3 12Bvision
google
131,072$0.050/M$0.150/M$0.075/M
Z.ai: GLM Flash Latestvisioncache
~z-ai
1,048,576$0.048/M$0.160/M$0.076/M
Z.ai: GLM 5.3 Flashvisioncache
z-ai
1,048,576$0.048/M$0.160/M$0.076/M
Z.ai: GLM 5.3 Flash (batch)visioncache
z-ai
1,048,576$0.050/M$0.180/M$0.083/M
Upstage: Solar Mini 4cache
upstage
524,288$0.050/M$0.200/M$0.087/M
Google: Gemini 2.5 Flash Lite (batch)visioncache
google
1,048,576$0.050/M$0.200/M$0.087/M
OpenAI: GPT-4.1 Nano (batch)visioncache
openai
1,047,576$0.050/M$0.200/M$0.087/M
inclusionAI: Ling 3.0 Flash VLvisioncache
inclusionai
131,072$0.060/M$0.180/M$0.090/M
Inference.net: Schematron V2 Smallcache
inference-net
128,000$0.050/M$0.230/M$0.095/M
Anthropic: Claude Haiku 5.5 (batch)visioncache
anthropic
1,000,000$0.050/M$0.250/M$0.100/M
OpenAI: GPT-6 Luna Pro (batch)visioncache
openai
1,050,000$0.050/M$0.250/M$0.100/M
OpenAI: GPT-6 Luna (batch)visioncache
openai
1,050,000$0.050/M$0.250/M$0.100/M
Mistral: Ministral 3 3B 2512visioncache
mistralai
131,072$0.100/M$0.100/M$0.100/M
NVIDIA: Nemotron 3.5 Lightningcache
nvidia
262,144$0.070/M$0.200/M$0.102/M
NVIDIA: Nemotron 3 Nano 30B A3B
nvidia
262,144$0.060/M$0.240/M$0.105/M
Amazon: Nova Lite 1.0vision
amazon
300,000$0.060/M$0.240/M$0.105/M
IBM: Granite 4.2 8Bcache
ibm-granite
131,072$0.060/M$0.250/M$0.107/M
inclusionAI: Ling 3.1 Flashcache
inclusionai
262,144$0.075/M$0.225/M$0.113/M
Poolside: Laguna S 2.1cache
poolside
1,048,576$0.090/M$0.180/M$0.113/M
Qwen: Qwen3.5-9Bvision
qwen
262,144$0.100/M$0.150/M$0.113/M
DeepSeek: DeepSeek V4 Flash 0423cache
deepseek
1,048,576$0.091/M$0.182/M$0.114/M
Qwen: Qwen3.5-Flashvision
qwen
1,000,000$0.065/M$0.260/M$0.114/M
Nex AGI: Nex-N2.5-Provisioncache
nex-agi
262,144$0.075/M$0.250/M$0.119/M
Google: Gemma 4 26B A4B visioncache
google
262,144$0.076/M$0.255/M$0.121/M
Meta: Llama 3.2 3B Instruct
meta-llama
131,072$0.051/M$0.335/M$0.122/M
Qwen: Qwen3 Coder 30B A3B Instruct
qwen
262,144$0.070/M$0.280/M$0.122/M
Meta: Muse Spark 1.3 Contributorvisioncache
meta
1,048,576$0.100/M$0.200/M$0.125/M
Meta: Muse Spark 1.2 Contributorvisioncache
meta
1,048,576$0.100/M$0.200/M$0.125/M
ByteDance: UI-TARS 7B visioncache
bytedance
128,000$0.100/M$0.200/M$0.125/M
Qwen: Qwen3 32B
qwen
131,072$0.080/M$0.280/M$0.130/M
Mistral: Mistral Small 4 (batch)visioncache
mistralai
262,144$0.075/M$0.300/M$0.131/M
ByteDance Seed: Seed 1.6 Flashvision
bytedance-seed
262,144$0.075/M$0.300/M$0.131/M
OpenAI: gpt-oss-safeguard-20bcache
openai
131,072$0.075/M$0.300/M$0.131/M
OpenAI: GPT-4o-mini (batch)visioncache
openai
128,000$0.075/M$0.300/M$0.131/M
Mistral: Mistral Small 3.2 24Bvision
mistralai
256,000$0.094/M$0.250/M$0.133/M
OpenAI: GPT-5 Nanovisioncache
openai
400,000$0.050/M$0.400/M$0.138/M
Qwen: Qwen3 30B A3B Instruct 2507
qwen
262,144$0.090/M$0.300/M$0.142/M
Tencent: Hy3cache
tencent
262,144$0.083/M$0.330/M$0.144/M
Z.ai: GLM 4.7 Flash
z-ai
200,000$0.060/M$0.400/M$0.145/M
Google: Gemma 3 27Bvisioncache
google
131,072$0.080/M$0.350/M$0.147/M
StepFun: Step 3.5 Flash
stepfun
262,144$0.100/M$0.300/M$0.150/M
Mistral: Ministral 3 8B 2512visioncache
mistralai
262,144$0.150/M$0.150/M$0.150/M
Meta: Llama 4 Scoutvision
meta-llama
1,310,720$0.100/M$0.300/M$0.150/M
Meta: Llama 3.3 70B Instruct
meta-llama
131,072$0.100/M$0.320/M$0.155/M
Upstage: Solar Pro 4cache
upstage
524,288$0.090/M$0.360/M$0.158/M
Inception: Mercury 3cache
inception
1,000,000$0.100/M$0.350/M$0.163/M
NVIDIA: Nemotron 3 Super
nvidia
262,144$0.080/M$0.450/M$0.172/M
NVIDIA: Nemotron 3.5 Super VLvisioncache
nvidia
1,048,576$0.100/M$0.400/M$0.175/M
Xiaomi: MiMo-V2.6-Flashvisioncache
xiaomi
1,050,000$0.140/M$0.280/M$0.175/M
Xiaomi: MiMo-V2.5visioncache
xiaomi
1,050,000$0.140/M$0.280/M$0.175/M
ByteDance Seed: Seed-2.0-Minivision
bytedance-seed
262,144$0.100/M$0.400/M$0.175/M
Google: Gemini 2.5 Flash Litevisioncache
google
1,048,576$0.100/M$0.400/M$0.175/M
OpenAI: GPT-4.1 Nanovisioncache
openai
1,047,576$0.100/M$0.400/M$0.175/M
DeepSeek: DeepSeek Flash Latestvisioncache
~deepseek
1,048,576$0.035/M$0.600/M$0.176/M
Meta: Llama Guard 4 12Bvision
meta-llama
163,840$0.180/M$0.180/M$0.180/M
PrismML: Ternary Bonsai 2 27Bvisioncache
prism-ml
262,144$0.075/M$0.500/M$0.181/M
DeepSeek: DeepSeek V4 Flash 0731cache
deepseek
1,048,576$0.130/M$0.389/M$0.195/M
Anthropic: Claude Haiku 5.5visioncache
anthropic
1,000,000$0.100/M$0.500/M$0.200/M
OpenAI: GPT-6 Luna Provisioncache
openai
1,050,000$0.100/M$0.500/M$0.200/M
OpenAI: GPT-6 Lunavisioncache
openai
1,050,000$0.100/M$0.500/M$0.200/M
OpenAI: GPT Luna Latestvisioncache
~openai
1,050,000$0.100/M$0.500/M$0.200/M
NVIDIA: Nemotron 3.5 Content Safetyvision
nvidia
131,072$0.200/M$0.200/M$0.200/M
Anthropic: Claude Haiku Latestvisioncache
~anthropic
1,000,000$0.100/M$0.500/M$0.200/M
Mistral: Ministral 3 14B 2512visioncache
mistralai
262,144$0.200/M$0.200/M$0.200/M
Qwen: Qwen3 235B A22B Instruct 2507
qwen
262,144$0.090/M$0.550/M$0.205/M
LiquidAI: LFM2.5-24B-A2Bcache
liquid
128,000$0.050/M$0.700/M$0.212/M
OpenAI: GPT-5.6 Luna Pro (batch)visioncache
openai
1,050,000$0.100/M$0.600/M$0.225/M
OpenAI: GPT-5.6 Luna (batch)visioncache
openai
1,050,000$0.100/M$0.600/M$0.225/M
Mistral: Codestral 2508 (batch)cache
mistralai
256,000$0.150/M$0.450/M$0.225/M
Qwen: Qwen3.8 Omni Flashvisioncache
qwen
1,000,000$0.150/M$0.470/M$0.230/M
Qwen: Qwen3.8 Flashvisioncache
qwen
1,000,000$0.150/M$0.470/M$0.230/M
OpenAI: GPT-5.4 Nano (batch)visioncache
openai
400,000$0.100/M$0.625/M$0.231/M
Tencent: Hunyuan A13B Instruct
tencent
131,072$0.140/M$0.570/M$0.247/M
Google: Gemma 4 31Bvision
google
262,144$0.200/M$0.400/M$0.250/M
DeepSeek: DeepSeek V4.1 Flash 1017visioncache
deepseek
1,048,576$0.150/M$0.600/M$0.263/M
Mistral: Mistral Small 4visioncache
mistralai
262,144$0.150/M$0.600/M$0.263/M
Upstage: Solar Pro 3cache
upstage
131,072$0.150/M$0.600/M$0.263/M
Qwen: Qwen3 VL 30B A3B Instructvision
qwen
262,144$0.150/M$0.600/M$0.263/M

Frequently asked questions

How many long context models are listed?
There are 401 models matching this filter right now, pulled live from the OpenRouter public API and updated daily.
How often does pricing change?
AI model prices change frequently — sometimes weekly. Token Price Index rebuilds this page daily from the OpenRouter API so you always see current prices. Subscribe for price-drop email alerts.
How is the blended cost calculated?
Blended = (3 × input cost + output cost) / 4. This weights input 3:1 over output, reflecting a typical read-heavy workload. Sort by input or output $/M if your workload differs.
How can I get alerted when a long context model price drops?
Token Price Index Pro emails you the moment any model you follow gets cheaper — for $9/mo. No more manually checking provider pricing pages.

Get price-drop alerts for models you use

$9/mo · cancel anytime

Pro subscribers get emailed the instant a model they track drops in price, plus a weekly cheapest-per-task digest and full CSV/JSON export.

Subscribe — $9/mo
Data: OpenRouter public model API, normalized to $/1M tokens. Blended = (3×input + output)/4. Generated 19 Oct 2026 22:00 UTC. Prices change often — confirm on the provider before committing spend.