EBEasy BenchmarksLLM model index
Workspace
Overview
Benchmarks
Benchmarks list
Overall Index
Coding
Math
MMLU-Pro
Speed
Value
Models
All models
GPT-5.5 (xhigh)
GPT-5.5 (high)
Claude Opus 4.7 (Adaptive Reasoning, Max Effort)
Gemini 3.1 Pro Preview
GPT-5.4 (xhigh)
Artificial Analysis data
Back

Output Price

Output token price per 1M tokens.

Output Price measures the listed cost of generated text. Output tokens are often priced higher than input tokens, so this metric can dominate workloads with long answers or generated artifacts.

Test type: Provider output-token price comparison. Lower values rank better.

Coverage

325 models have this metric.

$0.040/M

Current leader: Gemma 3n E4B Instruct

Project links

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

Artificial Analysis methodology

Top Output $ Models

Top models ranked by Output $.

Leaderboard

RankModelCreatorValueSpeedBlended Price
#1Gemma 3n E4B InstructGoogle$0.040/M15.3 tok/s$0.025/M
#2
Qwen3.5 0.8B (Non-reasoning)
Alibaba
$0.050/M
273.6 tok/s
$0.020/M
#3Qwen3.5 0.8B (Reasoning)Alibaba$0.050/Mn/a$0.020/M
#4Granite 4.1 8BIBM$0.100/M134.6 tok/s$0.063/M
#5Llama 3.1 Instruct 8BMeta$0.100/M164.4 tok/s$0.100/M
#6Llama 3.2 Instruct 1BMeta$0.100/M97.7 tok/s$0.100/M
#7Ministral 3 3BMistral$0.100/M287.6 tok/s$0.100/M
#8Qwen3.5 2B (Non-reasoning)Alibaba$0.100/M227 tok/s$0.040/M
#9Qwen3.5 2B (Reasoning)Alibaba$0.100/Mn/a$0.040/M
#10LFM2 24B A2BLiquid AI$0.120/M196.9 tok/s$0.052/M
#11Nova MicroAmazon$0.140/M332.1 tok/s$0.061/M
#12Llama 3 Instruct 8BMeta$0.145/M82.2 tok/s$0.070/M
#13Llama 3.2 Instruct 3BMeta$0.150/M52.2 tok/s$0.150/M
#14Ministral 3 8BMistral$0.150/M157.6 tok/s$0.150/M
#15Qwen3.5 4B (Non-reasoning)Alibaba$0.150/M200.2 tok/s$0.060/M
#16Qwen3.5 4B (Reasoning)Alibaba$0.150/M204.8 tok/s$0.060/M
#17Qwen3.5 9B (Reasoning)Alibaba$0.150/M62.9 tok/s$0.113/M
#18Solar MiniUpstage$0.150/M41.7 tok/s$0.150/M
#19NVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIA$0.160/M121.6 tok/s$0.070/M
#20NVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIA$0.195/M153.3 tok/s$0.086/M
#21Apertus 8B InstructSwiss AI Initiative$0.200/Mn/a$0.125/M
#22gpt-oss-20B (high)OpenAI$0.200/M242.3 tok/s$0.100/M
#23Ministral 3 14BMistral$0.200/M121.6 tok/s$0.200/M
#24Olmo 3 7B InstructAllen Institute for AI$0.200/Mn/a$0.125/M
#25Qwen2.5 TurboAlibaba$0.200/M77.7 tok/s$0.087/M
#26NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA$0.220/M154.8 tok/s$0.096/M
#27gpt-oss-20B (low)OpenAI$0.225/M249.7 tok/s$0.108/M
#28Nova LiteAmazon$0.240/M186.8 tok/s$0.105/M
#29Llama 3.2 Instruct 11B (Vision)Meta$0.245/M77.4 tok/s$0.245/M
#30Granite 3.3 8B (Non-reasoning)IBM$0.250/M410.5 tok/s$0.085/M
#31Granite 4.0 H SmallIBM$0.250/M238.9 tok/s$0.107/M
#32Llama 2 Chat 7BMeta$0.250/M99.7 tok/s$0.100/M
#33Mistral 7B InstructMistral$0.250/M156.9 tok/s$0.250/M
#34DeepSeek V4 Flash (Reasoning, High Effort)DeepSeek$0.280/Mn/a$0.175/M
#35DeepSeek V4 Flash (Reasoning, Max Effort)DeepSeek$0.280/M77.4 tok/s$0.175/M
#36Devstral Small (Jul '25)Mistral$0.300/M194.2 tok/s$0.150/M
#37Hermes 3 - Llama-3.1 70BNous Research$0.300/M28.8 tok/s$0.300/M
#38Ling 2.6 FlashInclusionAI$0.300/M206 tok/s$0.150/M
#39MiMo-V2-Flash (Feb 2026)Xiaomi$0.300/M120.6 tok/s$0.150/M
#40MiMo-V2-Flash (Non-reasoning)Xiaomi$0.300/M116.7 tok/s$0.150/M
#41MiMo-V2-Flash (Reasoning)Xiaomi$0.300/M118.8 tok/s$0.150/M
#42Mistral Small 3Mistral$0.300/M135.9 tok/s$0.150/M
#43Mistral Small 3.1Mistral$0.300/M138.8 tok/s$0.150/M
#44Mistral Small 3.2Mistral$0.300/M153.8 tok/s$0.150/M
#45Step 3.5 FlashStepFun$0.300/M123.6 tok/s$0.150/M
#46Gemini 2.5 Flash-Lite (Non-reasoning)Google$0.400/M239.9 tok/s$0.175/M
#47Gemini 2.5 Flash-Lite (Reasoning)Google$0.400/M243.6 tok/s$0.175/M
#48Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)Google$0.400/Mn/a$0.175/M
#49Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)Google$0.400/Mn/a$0.175/M
#50Gemma 4 26B A4B (Reasoning)Google$0.400/Mn/a$0.198/M
#51GLM-4.7-Flash (Non-reasoning)Z AI$0.400/M89.6 tok/s$0.152/M
#52GLM-4.7-Flash (Reasoning)Z AI$0.400/M110.5 tok/s$0.152/M
#53GPT-4.1 nanoOpenAI$0.400/M125.2 tok/s$0.175/M
#54GPT-5 nano (high)OpenAI$0.400/M136 tok/s$0.138/M
#55GPT-5 nano (medium)OpenAI$0.400/M150.3 tok/s$0.138/M
#56GPT-5 nano (minimal)OpenAI$0.400/M139.1 tok/s$0.138/M
#57Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research$0.400/M83.5 tok/s$0.198/M
#58Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research$0.400/M78.6 tok/s$0.198/M
#59Jamba 1.5 MiniAI21 Labs$0.400/Mn/a$0.250/M
#60Jamba 1.6 MiniAI21 Labs$0.400/M184.5 tok/s$0.250/M
#61Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIA$0.400/M51.3 tok/s$0.175/M
#62Llama Nemotron Super 49B v1.5 (Reasoning)NVIDIA$0.400/M50.8 tok/s$0.175/M
#63DeepSeek V3.2 (Non-reasoning)DeepSeek$0.420/Mn/a$0.315/M
#64DeepSeek V3.2 (Reasoning)DeepSeek$0.420/Mn/a$0.315/M
#65DeepSeek V3.2 Exp (Non-reasoning)DeepSeek$0.420/Mn/a$0.315/M
#66DeepSeek V3.2 Exp (Reasoning)DeepSeek$0.420/Mn/a$0.315/M
#67Qwen3 0.6B (Non-reasoning)Alibaba$0.420/M204.8 tok/s$0.188/M
#68Qwen3 1.7B (Non-reasoning)Alibaba$0.420/M139 tok/s$0.188/M
#69Qwen3 4B (Non-reasoning)Alibaba$0.420/M104.4 tok/s$0.188/M
#70Grok 3 mini Reasoning (high)xAI$0.500/M215.5 tok/s$0.350/M
#71Grok 4 Fast (Non-reasoning)xAI$0.500/M77.4 tok/s$0.275/M
#72Grok 4 Fast (Reasoning)xAI$0.500/M76.2 tok/s$0.275/M
#73Grok 4.1 Fast (Non-reasoning)xAI$0.500/M112.1 tok/s$0.275/M
#74Grok 4.1 Fast (Reasoning)xAI$0.500/M140.9 tok/s$0.275/M
#75Phi-4Microsoft$0.500/M41.6 tok/s$0.219/M
#76Llama 3.1 Instruct 70BMeta$0.560/M32.2 tok/s$0.560/M
#77Ling-flash-2.0InclusionAI$0.570/M87.3 tok/s$0.247/M
#78Ring-flash-2.0InclusionAI$0.570/M91 tok/s$0.247/M
#79Seed-OSS-36B-InstructByteDance Seed$0.570/M40 tok/s$0.300/M
#80Gemini 2.0 Flash (Feb '25)Google$0.600/Mn/a$0.263/M