LLM Leaderboard - Comparison of over 100 AI models from OpenAI, Google, DeepSeek & others
Comparison and ranking the performance of over 100 AI models (LLMs) across key metrics including intelligence, price, performance and speed (output speed - tokens per second & latency - TTFT), context window & others. For more details including relating to our methodology, see our FAQs.
HIGHLIGHTS
GPT-5.2 (xhigh) and 
Claude Opus 4.5 are the highest intelligence models, followed by 
Gemini 3 Pro Preview (high) & GPT-5.1 (high).Output Speed (tokens/s):
Granite 3.3 8B (470 t/s) and 
Nova Micro (468 t/s) are the fastest models, followed by Granite 4.0 H Small & o3.Latency (seconds):
Apriel-v1.5-15B-Thinker (0.21s) and Apriel-v1.6-15B-Thinker (0.21s) are the lowest latency models, followed by 
DeepSeek-OCR & 
Ministral 3B (Dec '25).Price ($ per M tokens): Gemma 3n E4B ($0.03) and DeepSeek-OCR ($0.05) are the cheapest models, followed by 
Llama 3.2 1B & Llama 3.2 3B.Context Window: Llama 4 Scout (10m) and 
Grok 4.1 Fast (2m) are the largest context window models, followed by Grok 4 Fast & Grok 4 Fast.Features | Intelligence | Price | Speed | Latency | |||
|---|---|---|---|---|---|---|---|
Further Analysis | |||||||
GPT-5.2 (xhigh) | 400k | OpenAI | 51 | $4.81 | 100 | 33.31 | |
Claude Opus 4.5 | 200k | Anthropic | 49 | $10.00 | 83 | 1.78 | |
Gemini 3 Pro Preview (high) | 1m | | 48 | $4.50 | 119 | 35.40 | |
GPT-5.1 (high) | 400k | OpenAI | 47 | $3.44 | 130 | 30.07 | |
Gemini 3 Flash | 1m | | 46 | $1.13 | 207 | 12.25 | |
GPT-5.2 (medium) | 400k | OpenAI | 45 | $4.81 | 0 | 0.00 | |
Claude Opus 4.5 | 200k | Anthropic | 43 | $10.00 | 76 | 2.11 | |
Claude 4.5 Sonnet | 1m | Anthropic | 42 | $6.00 | 84 | 1.88 | |
GLM-4.7 | 200k |  Z AI | 42 | $0.94 | 81 | 0.89 | |
GPT-5.1 Codex (high) | 400k | OpenAI | 42 | $3.44 | 241 | 11.87 | |
Grok 4 | 256k | xAI | 41 | $6.00 | 37 | 18.18 | |
DeepSeek V3.2 | 128k | DeepSeek | 41 | $0.32 | 30 | 1.21 | |
o3 | 200k | OpenAI | 41E | $3.50 | 303 | 10.10 | |
GPT-5 mini (high) | 400k | OpenAI | 41 | $0.69 | 72 | 109.26 | |
Gemini 3 Pro Preview (low) | 1m | | 41 | $4.50 | 126 | 4.17 | |
Kimi K2 Thinking | 256k |  Kimi | 40 | $1.07 | 63 | 1.00 | |
MiniMax-M2.1 | 205k |  MiniMax | 39 | $0.53 | 63 | 2.10 | |
MiMo-V2-Flash | 256k |  Xiaomi | 39 | $0.15 | 112 | 1.47 | |
Grok 4.1 Fast | 2m | xAI | 38 | $0.28 | 165 | 7.30 | |
GPT-5.1 Codex mini (high) | 400k | OpenAI | 38 | $0.69 | 147 | 11.30 | |
Claude 4.5 Sonnet | 1m | Anthropic | 37 | $6.00 | 74 | 1.92 | |
Claude 4.5 Haiku | 200k | Anthropic | 37 | $2.00 | 114 | 0.51 | |
KAT-Coder-Pro V1 | 256k |  KwaiKAT | 36 | $0.00 | 66 | 1.03 | |
MiniMax-M2 | 205k | MiniMax | 36 | $0.53 | 97 | 1.60 | |
Nova 2.0 Pro Preview (medium) | 256k | Amazon | 35 | $3.44 | 130 | 24.18 | |
Gemini 3 Flash | 1m | | 35 | $1.13 | 191 | 0.66 | |
Grok 4 Fast | 2m | xAI | 35 | $0.28 | 161 | 4.83 | |
Gemini 2.5 Pro | 1m | | 34 | $3.44 | 156 | 32.58 | |
DeepSeek V3.2 Speciale | 128k | DeepSeek | 34E | $0.42 | 0 | 0.00 | |
GLM-4.7 | 200k | Z AI | 34 | $0.94 | 78 | 0.74 | |
DeepSeek V3.1 Terminus | 128k | DeepSeek | 33 | $0.80 | 0 | 0.00 | |
Doubao Seed Code | 256k |  ByteDance Seed | 33 | $0.41 | 0 | 0.00 | |
GPT-5.2 | 400k | OpenAI | 33 | $4.81 | 73 | 0.57 | |
gpt-oss-120B (high) | 131k | OpenAI | 33 | $0.26 | 299 | 0.45 | |
Qwen3 Max Thinking | 262k |  Alibaba | 32 | $2.40 | 36 | 1.84 | |
Grok 3 mini Reasoning (high) | 1m | xAI | 32 | $0.35 | 154 | 0.82 | |
Nova 2.0 Pro Preview (low) | 256k | Amazon | 32 | $3.44 | 137 | 12.85 | |
K-EXAONE | 256k |  LG AI Research | 32 | $0.00 | 67 | 0.37 | |
DeepSeek V3.2 | 128k | DeepSeek | 32 | $0.32 | 28 | 1.06 | |
Qwen3 Max | 262k | Alibaba | 31 | $2.40 | 25 | 1.84 | |
Claude 4.5 Haiku | 200k | Anthropic | 30 | $2.00 | 107 | 0.49 | |
MiMo-V2-Flash | 256k | Xiaomi | 30 | $0.15 | 91 | 1.62 | |
Nova 2.0 Lite (medium) | 1m | Amazon | 30 | $0.85 | 248 | 17.07 | |
Qwen3 235B A22B 2507 | 256k | Alibaba | 29 | $2.63 | 64 | 1.39 | |
ERNIE 5.0 Thinking Preview | 128k |  Baidu | 29 | $1.47 | 0 | 0.00 | |
Qwen3 VL 32B | 256k | Alibaba | 29E | $2.63 | 42 | 2.11 | |
DeepSeek V3.1 Terminus | 128k | DeepSeek | 28 | $0.80 | 0 | 0.00 | |
Nova 2.0 Omni (medium) | 1m | Amazon | 28 | $0.85 | 0 | 0.00 | |
Kimi K2 0905 | 256k | Kimi | 28 | $1.20 | 59 | 0.88 | |
Apriel-v1.6-15B-Thinker | 128k | ServiceNow | 28 | $0.00 | 132 | 0.21 | |
Qwen3 VL 235B A22B | 262k | Alibaba | 27 | $2.63 | 36 | 1.25 | |
Magistral Medium 1.2 | 128k | Mistral | 27 | $2.75 | 32 | 0.85 | |
DeepSeek R1 0528 | 128k | DeepSeek | 27 | $2.36 | 0 | 0.00 | |
GPT-5 nano (high) | 400k | OpenAI | 27 | $0.14 | 137 | 125.21 | |
Qwen3 Next 80B A3B | 262k | Alibaba | 27 | $1.88 | 168 | 1.08 | |
Grok Code Fast 1 | 256k | xAI | 26E | $0.53 | 206 | 9.32 | |
Nova 2.0 Lite (low) | 1m | Amazon | 25 | $0.85 | 246 | 8.19 | |
Qwen3 Coder 480B | 262k | Alibaba | 25 | $3.00 | 40 | 1.80 | |
gpt-oss-20B (high) | 131k | OpenAI | 25 | $0.10 | 209 | 0.55 | |
NVIDIA Nemotron 3 Nano | 1m |  NVIDIA | 25 | $0.10 | 127 | 0.39 | |
Qwen3 235B 2507 | 256k | Alibaba | 24 | $1.23 | 49 | 1.03 | |
HyperCLOVA X SEED Think (32B) | 128k |  Naver | 24 | $0.00 | 0 | 0.00 | |
Motif-2-12.7B | 128k |  Motif Technologies | 24E | $0.00 | 0 | 0.00 | |
Nova 2.0 Omni (low) | 1m | Amazon | 24 | $0.85 | 0 | 0.00 | |
gpt-oss-120B (low) | 131k | OpenAI | 24 | $0.26 | 250 | 0.52 | |
GLM-4.6V | 128k | Z AI | 24 | $0.45 | 52 | 0.97 | |
Grok 4.1 Fast | 2m | xAI | 23 | $0.28 | 137 | 0.72 | |
GLM-4.5-Air | 128k | Z AI | 23 | $0.42 | 54 | 1.00 | |
Mi:dm K 2.5 Pro | 128k |  Korea Telecom | 23 | $0.00 | 0 | 0.00 | |
K-EXAONE | 256k | LG AI Research | 23 | $0.00 | 75 | 0.37 | |
Nova 2.0 Pro Preview | 256k | Amazon | 23 | $3.44 | 152 | 0.75 | |
Qwen3 4B 2507 | 262k | Alibaba | 23E | $0.00 | 0 | 0.00 | |
Grok 4 Fast | 2m | xAI | 23 | $0.28 | 143 | 0.63 | |
Qwen3 30B A3B 2507 | 262k | Alibaba | 23 | $0.75 | 168 | 1.13 | |
Magistral Small 1.2 | 128k | Mistral | 23E | $0.75 | 162 | 0.42 | |
Mistral Large 3 | 256k | Mistral | 22 | $0.75 | 48 | 0.56 | |
INTELLECT-3 | 131k |  Prime Intellect | 22 | $0.42 | 45 | 0.92 | |
Gemini 2.5 Flash-Lite (Sep) | 1m | | 22 | $0.17 | 0 | 0.00 | |
Devstral 2 | 256k | Mistral | 22 | $0.00 | 60 | 0.45 | |
Hermes 4 405B | 128k |  Nous Research | 22E | $1.50 | 31 | 2.13 | |
Mistral Medium 3.1 | 128k | Mistral | 21 | $0.80 | 95 | 0.40 | |
gpt-oss-20B (low) | 131k | OpenAI | 21 | $0.10 | 187 | 0.61 | |
K2-V2 (high) | 512k |  MBZUAI Institute of Foundation Models | 21 | $0.00 | 0 | 0.00 | |
Qwen3 VL 235B A22B | 262k | Alibaba | 21 | $1.23 | 33 | 1.19 | |
Ring-flash-2.0 | 128k |  InclusionAI | 21E | $0.25 | 51 | 1.78 | |
Hermes 4 70B | 128k | Nous Research | 20E | $0.20 | 74 | 1.06 | |
Qwen3 Next 80B A3B | 262k | Alibaba | 20 | $0.88 | 160 | 1.03 | |
Llama Nemotron Ultra | 128k | NVIDIA | 20E | $0.90 | 34 | 1.90 | |
Qwen3 Coder 30B A3B | 262k | Alibaba | 20 | $0.90 | 103 | 1.42 | |
Ling-flash-2.0 | 128k | InclusionAI | 20E | $0.25 | 33 | 1.72 | |
Gemini 2.5 Flash-Lite (Sep) | 1m | | 20 | $0.17 | 0 | 0.00 | |
Qwen3 VL 30B A3B | 256k | Alibaba | 20 | $0.75 | 102 | 0.88 | |
Ling-1T | 128k | InclusionAI | 19 | $0.00 | 0 | 0.00 | |
Devstral Small 2 | 256k | Mistral | 19 | $0.00 | 94 | 0.38 | |
Llama Nemotron Super 49B v1.5 | 128k | NVIDIA | 19 | $0.17 | 71 | 0.48 | |
Nova Premier | 1m | Amazon | 19 | $5.00 | 78 | 1.11 | |
K2-V2 (medium) | 512k | MBZUAI Institute of Foundation Models | 19 | $0.00 | 0 | 0.00 | |
Devstral Medium | 256k | Mistral | 19 | $0.80 | 105 | 0.51 | |
Llama 4 Maverick | 1m | Meta | 19 | $0.46 | 110 | 0.63 | |
Llama 3.3 Nemotron Super 49B | 128k | NVIDIA | 18E | $0.00 | 0 | 0.00 | |
Nova 2.0 Lite | 1m | Amazon | 18 | $0.85 | 226 | 0.64 | |
GLM-4.6V | 128k | Z AI | 17 | $0.45 | 48 | 1.24 | |
ERNIE 4.5 300B A47B | 131k | Baidu | 17E | $0.48 | 18 | 2.36 | |
Hermes 4 405B | 128k | Nous Research | 17E | $1.50 | 28 | 1.82 | |
Qwen3 VL 32B | 256k | Alibaba | 17 | $1.23 | 42 | 1.93 | |
EXAONE 4.0 32B | 131k | LG AI Research | 17 | $0.70 | 58 | 0.50 | |
Nova 2.0 Omni | 1m | Amazon | 17 | $0.85 | 226 | 0.77 | |
Qwen3 VL 8B | 256k | Alibaba | 17 | $0.66 | 58 | 0.98 | |
Olmo 3 7B Think | 66k |  Allen Institute for AI | 17E | $0.14 | 74 | 0.63 | |
DeepSeek R1 0528 Qwen3 8B | 33k | DeepSeek | 16E | $0.07 | 51 | 1.81 | |
Qwen3 VL 30B A3B | 256k | Alibaba | 16 | $0.35 | 95 | 1.05 | |
Ministral 14B (Dec '25) | 256k | Mistral | 16 | $0.20 | 101 | 0.41 | |
Qwen3 4B 2507 | 262k | Alibaba | 16E | $0.00 | 0 | 0.00 | |
Falcon-H1R-7B | 256k |  TII UAE | 16 | $0.00 | 0 | 0.00 | |
DeepSeek R1 Distill Llama 70B | 128k | DeepSeek | 16E | $0.88 | 34 | 1.62 | |
Qwen3 Omni 30B A3B | 66k | Alibaba | 16 | $0.43 | 45 | 2.08 | |
Devstral Small | 256k | Mistral | 15 | $0.15 | 207 | 0.36 | |
Solar Pro 2 | 66k |  Upstage | 15 | $0.00 | 0 | 0.00 | |
Qwen3 30B A3B 2507 | 262k | Alibaba | 15 | $0.35 | 75 | 1.18 | |
NVIDIA Nemotron Nano 9B V2 | 131k | NVIDIA | 15 | $0.07 | 105 | 0.44 | |
Llama Nemotron Super 49B v1.5 | 128k | NVIDIA | 15 | $0.17 | 65 | 0.53 | |
Ling-mini-2.0 | 131k | InclusionAI | 15E | $0.12 | 106 | 1.68 | |
K2-V2 (low) | 512k | MBZUAI Institute of Foundation Models | 15 | $0.00 | 0 | 0.00 | |
Mistral Small 3.2 | 128k | Mistral | 15 | $0.15 | 105 | 0.32 | |
Qwen3 VL 4B | 256k | Alibaba | 15E | $0.00 | 0 | 0.00 | |
Ministral 8B (Dec '25) | 256k | Mistral | 15 | $0.15 | 102 | 0.30 | |
Llama 3.3 70B | 128k | Meta | 15 | $0.64 | 98 | 0.68 | |
NVIDIA Nemotron Nano 12B v2 VL | 128k | NVIDIA | 15 | $0.30 | 112 | 0.33 | |
Qwen3 VL 8B | 256k | Alibaba | 15 | $0.31 | 101 | 2.10 | |
Llama 3.1 Nemotron Nano 4B v1.1 | 128k | NVIDIA | 14E | $0.00 | 0 | 0.00 | |
Kimi Linear 48B A3B Instruct | 1m | Kimi | 14E | $0.00 | 0 | 0.00 | |
Reka Flash 3 | 128k |  Reka AI | 14E | $0.35 | 45 | 1.76 | |
Llama 3.3 Nemotron Super 49B | 128k | NVIDIA | 14E | $0.00 | 0 | 0.00 | |
Llama 3.1 405B | 128k | Meta | 14 | $4.19 | 24 | 1.22 | |
Qwen3 VL 4B | 256k | Alibaba | 14E | $0.00 | 0 | 0.00 | |
Solar Pro 2 | 66k | Upstage | 14 | $0.00 | 0 | 0.00 | |
Llama 4 Scout | 10m | Meta | 14 | $0.28 | 148 | 0.55 | |
Llama 3.1 Nemotron 70B | 128k | NVIDIA | 14 | $1.20 | 37 | 1.33 | |
Hermes 4 70B | 128k | Nous Research | 14E | $0.20 | 68 | 0.95 | |
NVIDIA Nemotron 3 Nano | 1m | NVIDIA | 14 | $0.10 | 100 | 0.35 | |
Command A | 256k |  Cohere | 13 | $4.38 | 41 | 0.53 | |
NVIDIA Nemotron Nano 9B V2 | 131k | NVIDIA | 13 | $0.10 | 80 | 0.59 | |
Phi-4 | 16k |  Microsoft Azure | 13E | $0.22 | 6 | 2.67 | |
Qwen3 1.7B | 32k | Alibaba | 13E | $0.40 | 96 | 1.02 | |
Jamba Reasoning 3B | 262k |  AI21 Labs | 13E | $0.00 | 0 | 0.00 | |
R1 1776 | 128k |  Perplexity | 12E | $0.00 | 0 | 0.00 | |
Llama 3.2 90B (Vision) | 128k | Meta | 12E | $0.72 | 41 | 0.62 | |
Ministral 3B (Dec '25) | 256k | Mistral | 12 | $0.10 | 102 | 0.29 | |
EXAONE 4.0 32B | 131k | LG AI Research | 12 | $0.70 | 64 | 0.54 | |
Nova Micro | 130k | Amazon | 12E | $0.06 | 468 | 0.45 | |
LFM2 8B A1B | 33k |  Liquid AI | 11E | $0.00 | 0 | 0.00 | |
Granite 4.0 H Small | 128k | IBM | 11 | $0.11 | 388 | 14.48 | |
Granite 4.0 Micro | 128k | IBM | 11E | $0.00 | 0 | 0.00 | |
Phi-4 Mini | 128k | Microsoft Azure | 11E | $0.00 | 45 | 0.29 | |
DeepHermes 3 - Mistral 24B | 32k | Nous Research | 11E | $0.00 | 0 | 0.00 | |
Llama 3.2 11B (Vision) | 128k | Meta | 11E | $0.16 | 67 | 0.61 | |
Qwen3 Omni 30B A3B | 66k | Alibaba | 11 | $0.43 | 79 | 1.95 | |
Qwen3 0.6B | 32k | Alibaba | 11E | $0.40 | 134 | 0.89 | |
Gemma 3 27B | 128k | | 10 | $0.00 | 40 | 3.75 | |
NVIDIA Nemotron Nano 12B v2 VL | 128k | NVIDIA | 10 | $0.30 | 106 | 0.70 | |
Phi-4 Multimodal | 128k | Microsoft Azure | 10E | $0.00 | 18 | 0.52 | |
Gemma 3n E2B | 32k | | 10E | $0.00 | 43 | 0.43 | |
Jamba 1.7 Large | 256k | AI21 Labs | 9 | $3.50 | 34 | 1.13 | |
Molmo 7B-D | 4k | Allen Institute for AI | 9E | $0.00 | 0 | 0.00 | |
Gemma 3 12B | 128k | | 9 | $0.00 | 47 | 27.70 | |
Gemma 3 1B | 32k | | 9E | $0.00 | 36 | 0.51 | |
Exaone 4.0 1.2B | 64k | LG AI Research | 8 | $0.00 | 0 | 0.00 | |
Gemma 3 270M | 32k | | 8E | $0.00 | 0 | 0.00 | |
Exaone 4.0 1.2B | 64k | LG AI Research | 8 | $0.00 | 0 | 0.00 | |
Granite 4.0 H 1B | 128k | IBM | 8 | $0.00 | 0 | 0.00 | |
Olmo 3 7B | 66k | Allen Institute for AI | 8 | $0.13 | 25 | 0.75 | |
LFM2 2.6B | 33k | Liquid AI | 8 | $0.00 | 0 | 0.00 | |
DeepHermes 3 - Llama-3.1 8B | 128k | Nous Research | 8E | $0.00 | 0 | 0.00 | |
Jamba 1.7 Mini | 258k | AI21 Labs | 7 | $0.25 | 103 | 0.80 | |
Granite 4.0 1B | 128k | IBM | 7 | $0.00 | 0 | 0.00 | |
Granite 4.0 350M | 33k | IBM | 7 | $0.00 | 0 | 0.00 | |
Qwen3 1.7B | 32k | Alibaba | 7 | $0.19 | 92 | 0.96 | |
LFM2 1.2B | 33k | Liquid AI | 6 | $0.00 | 0 | 0.00 | |
Gemma 3 4B | 128k | | 6 | $0.00 | 43 | 1.14 | |
Gemma 3n E4B | 32k | | 6 | $0.03 | 51 | 0.55 | |
Qwen3 0.6B | 32k | Alibaba | 6 | $0.19 | 127 | 0.86 | |
Granite 4.0 H 350M | 33k | IBM | 6 | $0.00 | 0 | 0.00 | |
DeepSeek-OCR | 8k | DeepSeek | -- | $0.05 | 198 | 0.24 | |
Grok Voice Agent | 32k | xAI | -- | $0.00 | 0 | 0.00 | |
Olmo 3.1 32B Think | 66k | Allen Institute for AI | -- | $0.00 | 57 | 0.96 | |
Cogito v2.1 | 128k |  Deep Cogito | -- | $1.25 | 56 | 0.49 | |
Mi:dm K 2.5 Pro Preview | 128k | Korea Telecom | -- | $0.00 | 0 | 0.00 | |
Doubao-Seed-1.8 | 256k | ByteDance Seed | -- | $0.15 | 0 | 0.00 | |