| Model | Provider | |||||||
|---|---|---|---|---|---|---|---|---|
MythoMax 13B gryphe | Fireworks | - | $0.07 | 0.22s | 108.8T/s | 99.95% | ||
Gemini Flash 1.5 google | Google Vertex | #28 | $0.08 | 2.81s | 172.4T/s | 99.86% | ||
GPT-4o-mini openai | OpenAI | #23 | $0.15 | 0.65s | 76.3T/s | 97.97% | ||
Llama 3.1 70B Instruct meta-llama | Together | #39 | $0.12 | 0.43s | 83.6T/s | 100.00% | ||
Llama 3.2 1B Instruct meta-llama | Lepton | #149 | $0.01 | 0.31s | 298.6T/s | 100.00% | ||
Gemini Flash 1.5 8B google | Google AI Studio | #67 | $0.04 | 0.48s | 222.9T/s | 99.99% | ||
DeepSeek V3 deepseek | DeepSeek | #8 | $0.14 | 1.13s | 46.4T/s | 99.53% | ||
Mistral Tiny mistralai | Mistral | #138 | $0.25 | 1.23s | 50.9T/s | 100.00% | ||
GPT-4o-mini (2024-07-18) openai | OpenAI | #23 | $0.15 | 0.40s | 64.2T/s | 100.00% | ||
Mistral Nemo mistralai | Mistral | - | $0.04 | 1.36s | 79.5T/s | 99.99% | ||
Llama 3.1 8B Instruct meta-llama | Avian.io | #103 | $0.02 | 0.48s | 272.3T/s | 99.99% | ||
Llama 3.3 70B Instruct meta-llama | Avian.io | #23 | $0.12 | 0.50s | 88.1T/s | 99.97% | ||
OpenChat 3.5 7B openchat | Lepton | #121 | $0.06 | 0.40s | 83.4T/s | 100.00% | ||
WizardLM-2 7B microsoft | Lepton | - | $0.06 | 0.24s | 56.0T/s | 100.00% | ||
WizardLM-2 8x22B microsoft | Together | - | $0.50 | 0.46s | 66.0T/s | 100.00% | ||
Llama 3 70B Instruct meta-llama | Fireworks | #56 | $0.23 | 0.50s | 160.8T/s | 100.00% | ||
Llama 3 8B Instruct meta-llama | Fireworks | #94 | $0.03 | 0.26s | 213.1T/s | 99.98% | ||
Mistral 7B Instruct v0.3 mistralai | Together | - | $0.03 | 0.14s | 153.2T/s | 96.00% | ||
Llama 3 Euryale 70B v2.1 sao10k | DeepInfra | - | $0.71 | 0.41s | 32.0T/s | 99.99% | ||
Hermes 3 405B Instruct nousresearch | Lambda | #15 | $0.81 | 1.24s | 18.5T/s | 100.00% | ||
Command R (08-2024) cohere | Cohere | #69 | $0.15 | 0.25s | 90.4T/s | 100.00% | ||
Qwen2.5 72B Instruct qwen | Together | #38 | $0.23 | 1.61s | 93.6T/s | 100.00% | ||
Llama 3.2 3B Instruct meta-llama | Lambda | #121 | $0.02 | 0.34s | 356.6T/s | 100.00% | ||
Ministral 8B mistralai | Mistral | #82 | $0.10 | 1.25s | 60.7T/s | 100.00% | ||
Claude 3.5 Sonnet anthropic | Anthropic | #7 | $3.12 | 2.42s | 57.3T/s | 99.90% | ||
Qwen2.5 Coder 32B Instruct qwen | Fireworks | #56 | $0.07 | 0.64s | 95.8T/s | 100.00% | ||
Mistral Large 2411 mistralai | Mistral | #40 | $2.05 | 1.69s | 46.0T/s | 95.99% | ||
GPT-3.5 Turbo 16k openai | OpenAI | - | $0.51 | 0.31s | 123.5T/s | 100.00% | ||
GPT-3.5 Turbo openai | OpenAI | #99 | $0.51 | 0.34s | 127.4T/s | 99.97% | ||
Hermes 13B nousresearch | NovitaAI | - | $0.17 | 0.59s | 81.5T/s | 100.00% | ||
OpenHermes 2.5 Mistral 7B teknium | NovitaAI | #122 | $0.17 | 1.04s | 61.6T/s | 99.93% | ||
Mixtral 8x7B Instruct mistralai | Fireworks | #113 | $0.24 | 0.38s | 142.7T/s | 100.00% | ||
Gemini Pro 1.5 google | Google AI Studio | #9 | $1.29 | 0.83s | 53.7T/s | 100.00% | ||
Mixtral 8x22B Instruct mistralai | Together | #89 | $0.91 | 0.35s | 67.9T/s | 100.00% | ||
Llama 3 8B Instruct (extended) meta-llama | Mancer (private) | - | $0.20 | 0.47s | 33.9T/s | 100.00% | ||
Hermes 2 Pro - Llama-3 8B nousresearch | NovitaAI | - | $0.03 | 0.63s | 145.2T/s | 100.00% | ||
Mistral 7B Instruct (nitro) mistralai | Lepton | - | $0.07 | 0.25s | 69.1T/s | 68.00% | ||
Mistral 7B Instruct mistralai | Together | #138 | $0.03 | 0.27s | 171.7T/s | 100.00% | ||
Mistral 7B Instruct (free) mistralai | Lepton | - | $0.00 | 0.36s | 84.5T/s | 100.00% | ||
Dolphin 2.9.2 Mixtral 8x22B 🐬 cognitivecomputations | NovitaAI | - | $0.91 | 6.39s | 9.2T/s | 99.65% | ||
Gemma 2 27B google | Together | #46 | $0.27 | 0.37s | 64.6T/s | 99.97% | ||
Llama 3.1 405B Instruct meta-llama | Fireworks | - | $0.81 | 0.87s | 53.8T/s | 99.97% | ||
GPT-4o (2024-08-06) openai | Azure | #16 | $2.58 | 3.28s | 128.4T/s | 100.00% | ||
Hermes 3 70B Instruct nousresearch | Hyperbolic | #43 | $0.12 | 0.98s | 31.9T/s | 100.00% | ||
Llama 3.1 Euryale 70B v2.2 sao10k | DeepInfra | - | $0.71 | 0.40s | 36.4T/s | 99.98% | ||
Command R+ (08-2024) cohere | Cohere | #48 | $2.45 | 0.38s | 59.0T/s | 99.95% | ||
Llama 3.2 1B Instruct (free) meta-llama | SambaNova | #149 | $0.00 | 0.70s | 2148.5T/s | 98.77% | ||
Rocinante 12B thedrummer | Infermatic | - | $0.25 | 0.83s | 22.7T/s | 100.00% | ||
Ministral 3B mistralai | Mistral | - | $0.04 | 1.19s | 94.1T/s | 99.99% | ||
Claude 3.5 Sonnet (self-moderated) anthropic | Anthropic | - | $3.12 | 2.52s | 59.2T/s | 99.18% | ||
Claude 3.5 Haiku anthropic | Google Vertex | #22 | $0.83 | 2.82s | 64.8T/s | 100.00% | ||
Unslopnemo 12b thedrummer | Infermatic | - | $0.50 | 0.55s | 68.9T/s | 99.50% | ||
GPT-4o (2024-11-20) openai | OpenAI | #1 | $2.58 | 0.72s | 122.5T/s | 99.98% | ||
Gemini Flash 2.0 Experimental (free) google | Google Vertex | #5 | $0.00 | 1.29s | 130.6T/s | 72.22% | ||
Grok 2 1212 x-ai | xAI | - | $2.08 | 0.50s | 62.3T/s | 99.68% | ||
Llama 3.3 Euryale 70B sao10k | Infermatic | - | $1.51 | 20.70s | 10.1T/s | 100.00% | ||
MythoMax 13B (nitro) gryphe | Fireworks | - | $0.20 | 0.22s | 108.8T/s | 100.00% | ||
ReMM SLERP 13B undi95 | Mancer (private) | - | $0.81 | 1.47s | 27.0T/s | 100.00% | ||
Mistral Small mistralai | Mistral | - | $0.20 | 1.51s | 51.3T/s | 100.00% | ||
Mistral Large mistralai | Mistral | #28 | $2.05 | 1.54s | 28.9T/s | 99.92% | ||
Claude 3 Haiku anthropic | Google Vertex | - | $0.26 | 1.00s | 148.1T/s | 100.00% | ||
Claude 3 Haiku (self-moderated) anthropic | Anthropic | - | $0.26 | 0.73s | 132.6T/s | 99.62% | ||
Command R cohere | Cohere | - | $0.49 | 0.08s | 119.4T/s | 100.00% | ||
Llama 3 8B Instruct (free) meta-llama | Together (lite) | - | $0.00 | 0.49s | 154.0T/s | 100.00% | ||
GPT-4o (extended) openai | OpenAI | - | $6.14 | 0.53s | 110.2T/s | 80.00% | ||
GPT-4o openai | OpenAI | - | $2.58 | 0.53s | 110.2T/s | 99.98% | ||
LlamaGuard 2 8B meta-llama | Together | - | $0.18 | 0.25s | 41.0T/s | 100.00% | ||
Claude 3.5 Sonnet (2024-06-20) anthropic | Anthropic | - | $3.12 | 1.32s | 59.7T/s | 99.84% | ||
Llama 3.1 70B Instruct (free) meta-llama | SambaNova | - | $0.00 | 1.66s | 363.1T/s | 80.00% | ||
Llama 3.1 Sonar 70B Online perplexity | Perplexity | #132 | $1.01 | 1.44s | 56.4T/s | 100.00% | ||
ChatGPT-4o openai | OpenAI | - | $5.12 | 0.52s | 122.0T/s | 99.88% | ||
o1-mini openai | OpenAI | #14 | $3.10 | 3.90s | 80.5T/s | 99.84% | ||
Llama 3.2 11B Vision Instruct meta-llama | Together | - | $0.06 | 1.01s | 118.8T/s | 99.98% | ||
Llama 3.1 Nemotron 70B Instruct nvidia | Together | #39 | $0.12 | 0.59s | 75.7T/s | 99.93% | ||
LearnLM 1.5 Pro Experimental (free) google | Google AI Studio | - | $0.00 | 0.57s | 58.8T/s | 99.99% | ||
Nova Pro 1.0 amazon | Amazon BedRock | - | $0.83 | 0.11s | 63.8T/s | 100.00% | ||
Command R7B (12-2024) cohere | Cohere | - | $0.04 | 0.10s | 102.9T/s | 98.42% | ||
Phi 4 microsoft | DeepInfra | - | $0.07 | 0.31s | 79.3T/s | 99.83% |
Created by JC, data from OpenRouter and Chatbot Arena