Easy Benchmarks
Workspace
Overview
Benchmarks
Benchmarks list
Compare
Overall Index
Coding
Math
MMLU-Pro
Speed
Value
LLMs
Audio
Image
Video
Feedback
Log inSign up
Back

Blended Price

Price per 1M tokens using the Artificial Analysis 3:1 blended ratio.

Blended Price is a cost metric for comparing providers with different input and output token prices. Artificial Analysis calculates it with a 3:1 input-to-output token ratio so a single number can summarize typical API cost.

Test type: Provider price normalization. Lower values rank better.

Coverage

357 models have this metric.

$0.020/M

Current leader: Qwen3.5 0.8B (Non-reasoning)

Project links

Prices come from Artificial Analysis pricing data in the committed snapshot.

Artificial Analysis methodology

Top Blended $ Models

Top models ranked by Blended $.

Leaderboard

RankModelCreatorValueSpeedBlended Price
#1Qwen3.5 0.8B (Non-reasoning)Alibaba$0.020/M88 tok/s$0.020/M
#2
Qwen3.5 0.8B (Reasoning)
Alibaba
$0.020/M
n/a
$0.020/M
#3Gemma 3n E4B InstructGoogle$0.025/M50 tok/s$0.025/M
#4Qwen3.5 2B (Non-reasoning)Alibaba$0.040/M318.9 tok/s$0.040/M
#5Qwen3.5 2B (Reasoning)Alibaba$0.040/Mn/a$0.040/M
#6Sarvam 30B (high)Sarvam$0.047/M147.9 tok/s$0.047/M
#7Gemma 3 4B InstructGoogle$0.050/Mn/a$0.050/M
#8Llama 3.2 Instruct 1BMeta$0.050/M92.9 tok/s$0.050/M
#9LFM2 24B A2BLiquid AI$0.052/M127.6 tok/s$0.052/M
#10Qwen3.5 4B (Non-reasoning)Alibaba$0.060/M208.9 tok/s$0.060/M
#11Qwen3.5 4B (Reasoning)Alibaba$0.060/M195.8 tok/s$0.060/M
#12Nova MicroAmazon$0.061/M284.2 tok/s$0.061/M
#13Granite 4.1 8BIBM$0.063/M134.2 tok/s$0.063/M
#14Llama 3 Instruct 8BMeta$0.070/M88.3 tok/s$0.070/M
#15NVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIA$0.070/M118 tok/s$0.070/M
#16Sarvam 105B (high)Sarvam$0.074/M100.7 tok/s$0.074/M
#17Granite 3.3 8B (Non-reasoning)IBM$0.085/M400.9 tok/s$0.085/M
#18NVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIA$0.086/M133.6 tok/s$0.086/M
#19gpt-oss-20B (high)OpenAI$0.088/M240 tok/s$0.088/M
#20NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIA$0.088/M87.3 tok/s$0.088/M
#21Qwen2.5 TurboAlibaba$0.088/M66.4 tok/s$0.088/M
#22gpt-oss-20B (low)OpenAI$0.095/M224.2 tok/s$0.095/M
#23NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA$0.096/M133.6 tok/s$0.096/M
#24Llama 2 Chat 7BMeta$0.100/M100.6 tok/s$0.100/M
#25Llama 3.1 Instruct 8BMeta$0.100/M201.5 tok/s$0.100/M
#26Ministral 3 3BMistral$0.100/M174.3 tok/s$0.100/M
#27Mistral Small 3Mistral$0.104/M153.7 tok/s$0.104/M
#28Nova LiteAmazon$0.105/M191.7 tok/s$0.105/M
#29Granite 4.0 H SmallIBM$0.107/M454.2 tok/s$0.107/M
#30Qwen3.5 9B (Reasoning)Alibaba$0.113/M69.4 tok/s$0.113/M
#31Apertus 8B InstructSwiss AI Initiative$0.125/Mn/a$0.125/M
#32Olmo 3 7B InstructAllen Institute for AI$0.125/Mn/a$0.125/M
#33Mistral Small 3.2Mistral$0.128/M127.1 tok/s$0.128/M
#34Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA$0.131/M276.7 tok/s$0.131/M
#35Qwen3 30B A3B (Non-reasoning)Alibaba$0.133/M68.6 tok/s$0.133/M
#36GPT-5 nano (high)OpenAI$0.138/M150.4 tok/s$0.138/M
#37GPT-5 nano (medium)OpenAI$0.138/M167 tok/s$0.138/M
#38GPT-5 nano (minimal)OpenAI$0.138/M153.9 tok/s$0.138/M
#39Mistral Small 3.1Mistral$0.138/M158.2 tok/s$0.138/M
#40Gemma 3 12B InstructGoogle$0.140/Mn/a$0.140/M
#41Gemma 3 27B InstructGoogle$0.145/Mn/a$0.145/M
#42Devstral Small (Jul '25)Mistral$0.150/M42.3 tok/s$0.150/M
#43Gemma 4 12B (Reasoning)Google$0.150/M158.6 tok/s$0.150/M
#44Ling 2.6 FlashInclusionAI$0.150/Mn/a$0.150/M
#45Llama 3.2 Instruct 3BMeta$0.150/M52.3 tok/s$0.150/M
#46MiMo-V2-Flash (Feb 2026)Xiaomi$0.150/M124.9 tok/s$0.150/M
#47MiMo-V2-Flash (Non-reasoning)Xiaomi$0.150/M122.8 tok/s$0.150/M
#48MiMo-V2-Flash (Reasoning)Xiaomi$0.150/M129.5 tok/s$0.150/M
#49Ministral 3 8BMistral$0.150/M103.6 tok/s$0.150/M
#50Solar MiniUpstage$0.150/M75.9 tok/s$0.150/M
#51Step 3.5 FlashStepFun$0.150/M217.5 tok/s$0.150/M
#52Step 3.5 Flash 2603StepFun$0.150/M231 tok/s$0.150/M
#53GLM-4.7-Flash (Non-reasoning)Z AI$0.153/M105.6 tok/s$0.153/M
#54GLM-4.7-Flash (Reasoning)Z AI$0.153/M94.1 tok/s$0.153/M
#55DeepSeek V4 Flash (Non-reasoning)DeepSeek$0.175/M97.4 tok/s$0.175/M
#56DeepSeek V4 Flash (Reasoning, High Effort)DeepSeek$0.175/Mn/a$0.175/M
#57DeepSeek V4 Flash (Reasoning, Max Effort)DeepSeek$0.175/M98.3 tok/s$0.175/M
#58Gemini 2.5 Flash-Lite (Non-reasoning)Google$0.175/M229.5 tok/s$0.175/M
#59Gemini 2.5 Flash-Lite (Reasoning)Google$0.175/M265.2 tok/s$0.175/M
#60Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)Google$0.175/Mn/a$0.175/M
#61Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)Google$0.175/Mn/a$0.175/M
#62GPT-4.1 nanoOpenAI$0.175/M118.2 tok/s$0.175/M
#63Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIA$0.175/M43.9 tok/s$0.175/M
#64Llama Nemotron Super 49B v1.5 (Reasoning)NVIDIA$0.175/M44.2 tok/s$0.175/M
#65MiMo-V2.5Xiaomi$0.175/M77.4 tok/s$0.175/M
#66Qwen3 30B A3B (Reasoning)Alibaba$0.180/M68.5 tok/s$0.180/M
#67Qwen3 8B (Non-reasoning)Alibaba$0.185/M64.4 tok/s$0.185/M
#68Qwen3 0.6B (Non-reasoning)Alibaba$0.188/Mn/a$0.188/M
#69Qwen3 1.7B (Non-reasoning)Alibaba$0.188/Mn/a$0.188/M
#70Qwen3 4B (Non-reasoning)Alibaba$0.188/Mn/a$0.188/M
#71Gemma 4 26B A4B (Non-reasoning)Google$0.198/M48 tok/s$0.198/M
#72Gemma 4 26B A4B (Reasoning)Google$0.198/Mn/a$0.198/M
#73Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research$0.198/M84.9 tok/s$0.198/M
#74Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research$0.198/M87.2 tok/s$0.198/M
#75TEHy3-preview (Non-reasoning)Tencent$0.200/M83.9 tok/s$0.200/M
#76TEHy3-preview (Reasoning)Tencent$0.200/M96 tok/s$0.200/M
#77Ministral 3 14BMistral$0.200/M106.9 tok/s$0.200/M
#78Gemma 4 31B (Non-reasoning)Google$0.205/M54.9 tok/s$0.205/M
#79Mistral 7B InstructMistral$0.206/M110.4 tok/s$0.206/M
#80Qwen3 30B A3B 2507 InstructAlibaba$0.213/M105.2 tok/s$0.213/M