Model Leaderboard
Compare key performance metrics for LLM APIs.
Updated at: 2/10/2026, 12:02:08 PM
| Model | Provider | |||||||
|---|---|---|---|---|---|---|---|---|
Gemini 2.5 Flash google | Google Vertex | - | $0.32 | 1.90s | 122.0T/s | 99.66% | ||
GPT-4o-mini openai | OpenAI | - | $0.15 | 0.46s | 21.0T/s | 99.99% | ||
Llama 3.1 8B Instruct meta-llama | Groq | - | $0.02 | 0.14s | 323.0T/s | 99.98% | ||
Gemini 2.0 Flash google | Google AI Studio | - | $0.10 | 0.47s | 77.0T/s | 96.82% | ||
Gemini 2.5 Flash Lite google | Google Vertex | - | $0.10 | 0.46s | 90.0T/s | 99.56% | ||
gpt-oss-120b openai | Cerebras | - | $0.04 | 0.40s | 728.0T/s | 99.65% | ||
Grok 4.1 Fast x-ai | xAI | - | $0.20 | 3.56s | 87.0T/s | 100.00% | ||
DeepSeek V3.2 deepseek | Google Vertex | - | $0.25 | 0.80s | 32.5T/s | 99.80% | ||
Gemini 3 Flash Preview google | Google AI Studio | - | $0.52 | 1.05s | 75.0T/s | 98.70% | ||
Mistral Nemo mistralai | Mistral | - | $0.02 | 0.34s | 75.0T/s | 100.00% | ||
Llama 3.3 70B Instruct meta-llama | Groq | - | $0.10 | 0.26s | 220.0T/s | 99.92% | ||
DeepSeek V3 deepseek-ai | Chutes | - | $0.31 | 0.76s | 41.0T/s | 99.92% | ||
Gemini 2.0 Flash Lite google | Google AI Studio | - | $0.08 | 0.39s | 95.0T/s | 99.35% | ||
Gemma 3 27B google | Phala | - | $0.04 | 0.76s | 33.0T/s | 91.03% | ||
Gemma 3 12B google | Cloudflare | - | $0.03 | 0.22s | 27.0T/s | 99.55% | ||
Gemma 3 4B google | Chutes | - | $0.02 | 0.45s | 37.0T/s | 99.98% | ||
DeepSeek V3 0324 deepseek | Nebius AI Studio (Fast) | - | $0.20 | 1.06s | 100.0T/s | 99.96% | ||
Llama 4 Maverick meta-llama | Groq | - | $0.15 | 0.27s | 138.0T/s | 99.91% | ||
GPT-4.1 Mini openai | Azure | - | $0.41 | 0.70s | 48.0T/s | 99.99% | ||
GPT-4.1 openai | Azure | - | $2.06 | 0.75s | 47.0T/s | 99.98% | ||
Qwen3 32B qwen | Cerebras | - | $0.08 | 0.40s | 608.0T/s | 99.98% | ||
Gemini 2.5 Pro google | Google AI Studio | - | $1.33 | 3.12s | 105.0T/s | 99.17% | ||
Mistral Small 3.2 24B mistralai | Mistral | - | $0.06 | 0.30s | 89.0T/s | 99.85% | ||
Qwen3 235B A22B Instruct 2507 qwen | Cerebras | - | $0.07 | 0.33s | 135.0T/s | 99.93% | ||
gpt-oss-20b openai | Groq | - | $0.03 | 0.11s | 494.0T/s | 99.59% | ||
GPT-5 Nano openai | OpenAI | - | $0.05 | 3.12s | 90.0T/s | 99.89% | ||
GPT-5 Mini openai | Azure | - | $0.27 | 9.08s | 81.0T/s | 99.86% | ||
DeepSeek V3.1 deepseek | Google Vertex | - | $0.16 | 0.98s | 92.0T/s | 99.89% | ||
Grok Code Fast 1 x-ai | xAI | - | $0.21 | 3.76s | 113.0T/s | 92.00% | ||
Qwen3 Next 80B A3B Thinking qwen | Hyperbolic | - | $0.16 | 0.81s | 227.0T/s | 98.21% | ||
Grok 4 Fast x-ai | xAI | - | $0.20 | 2.29s | 110.0T/s | 100.00% | ||
Qwen3 VL 235B A22B Instruct qwen | Chutes | - | $0.21 | 2.11s | 76.0T/s | 99.96% | ||
Gemini 2.5 Flash Lite Preview 09-2025 google | Google AI Studio | - | $0.10 | 1.16s | 319.0T/s | 99.90% | ||
Claude Sonnet 4.5 anthropic | Google Vertex (Global) | - | $3.12 | 1.55s | 39.0T/s | 99.95% | ||
Claude Haiku 4.5 anthropic | Anthropic | - | $1.04 | 0.76s | 64.0T/s | 99.99% | ||
Gemini 3 Pro Preview google | Google AI Studio | - | $2.10 | 3.46s | 78.0T/s | 99.50% | ||
GPT-5.2 openai | OpenAI | - | $1.86 | 2.36s | 35.0T/s | 99.86% | ||
GLM 4.7 z-ai | Cerebras | - | $0.41 | 0.64s | 154.0T/s | 99.81% | ||
MiniMax M2.1 minimax | Fireworks | - | $0.28 | 0.78s | 104.0T/s | 99.94% | ||
Kimi K2.5 moonshotai | Baseten | - | $0.47 | 0.74s | 96.0T/s | 99.81% | ||
Trinity Large Preview (free) arcee-ai | Arcee AI | - | $0.00 | 0.48s | 20.0T/s | 88.00% | ||
Claude Opus 4.6 anthropic | Google Vertex | - | $5.20 | 1.49s | 42.0T/s | 99.46% | ||
Pony Alpha openrouter | Stealth | - | $0.00 | 9.73s | 19.0T/s | - | ||
GPT-4o openai | Azure | - | $2.58 | 0.88s | 43.0T/s | 99.99% | ||
Gemma 2 9B google | Nebius AI Studio (Fast) | - | $0.03 | 0.25s | 95.0T/s | - | ||
Llama 3.1 70B Instruct meta-llama | DeepInfra (Turbo) | - | $0.40 | 0.16s | 21.0T/s | 99.99% | ||
Ministral 3B mistralai | Mistral | - | $0.04 | 0.23s | 73.0T/s | - | ||
Llama 3.1 Nemotron Ultra 253B v1 nvidia | Nebius Token Factory | - | $0.61 | 0.14s | 20.0T/s | - | ||
GPT-4.1 Nano openai | Azure | - | $0.10 | 0.59s | 79.5T/s | 99.99% | ||
Gemma 3n 4B google | Together | - | $0.02 | 0.19s | 26.0T/s | 100.00% | ||
Claude Sonnet 4 anthropic | Anthropic | - | $3.12 | 1.31s | 47.0T/s | 99.78% | ||
R1 0528 deepseek | Nebius AI Studio (Fast) | - | $0.41 | 0.91s | 130.5T/s | 99.90% | ||
Grok 3 Mini x-ai | xAI Fast | - | $0.30 | 5.44s | 73.0T/s | 99.96% | ||
DeepSeek R1T2 Chimera (free) tngtech | Chutes | - | $0.00 | 1.97s | 11.0T/s | 99.98% | ||
GLM 4 32B z-ai | Z.ai | - | $0.10 | 1.11s | 3.0T/s | - | ||
Kimi K2 0905 moonshotai | Groq | - | $0.41 | 0.25s | 136.0T/s | 99.97% | ||
DeepSeek V3.1 Terminus deepseek | SiliconFlow | - | $0.22 | 2.65s | 55.0T/s | 99.89% | ||
DeepSeek V3.2 Exp deepseek | SiliconFlow | - | $0.27 | 2.24s | 36.0T/s | 99.95% | ||
Claude Opus 4.5 anthropic | Amazon Bedrock | - | $5.20 | 2.31s | 41.0T/s | 99.92% | ||
MiMo-V2-Flash xiaomi | AtlasCloud | - | $0.09 | 0.88s | 82.0T/s | 99.98% | ||
Step 3.5 Flash (free) stepfun | StepFun | - | $0.00 | 1.54s | 133.0T/s | - |