Home / Compare
LLM model comparison
Side-by-side token counts and API cost for popular model pairs. Every comparison uses the same TokenCALC pricing catalog. Then open the calculator with your own prompt.
Main calculator · Cost calculator · Tokenizer · Providers · Pricing guide
How to use these comparisons
Each pair shows curated input rates and a sample support-chat cost. Treat sample costs as a starting point. Paste your real prompt into the cost calculator with either model preselected before you lock a budget.
Featured comparisons (102)
- GPT-5 vs Claude Opus 4.8$1.25/1M vs $5.00/1M input
- GPT-5 vs Gemini 3.1 Pro$1.25/1M vs $2.00/1M input
- Claude Opus 4.8 vs Gemini 3.1 Pro$5.00/1M vs $2.00/1M input
- GPT-4.1 vs Claude Sonnet 5$2.00/1M vs $2.00/1M input
- GPT-4.1 vs Gemini 2.5 Pro$2.00/1M vs $1.25/1M input
- Claude Sonnet 5 vs Gemini 2.5 Pro$2.00/1M vs $1.25/1M input
- GPT-5 Mini vs Claude Sonnet 4.6$0.250/1M vs $3.00/1M input
- GPT-4.1 Mini vs Claude Haiku 4.5$0.400/1M vs $1.00/1M input
- GPT-4o mini vs Gemini 2.5 Flash$0.150/1M vs $0.300/1M input
- Gemini 2.5 Flash vs Claude Haiku 4.5$0.300/1M vs $1.00/1M input
- Gemini 3 Flash vs GPT-5 Mini$0.500/1M vs $0.250/1M input
- GPT-4.1 Nano vs Gemini 2.5 Flash-Lite$0.100/1M vs $0.100/1M input
- Claude Haiku 4.5 vs Gemini 2.5 Flash-Lite$1.00/1M vs $0.100/1M input
- DeepSeek V4 Flash vs GPT-4.1 Nano$0.140/1M vs $0.100/1M input
- DeepSeek V4 Flash vs Gemini 3.1 Flash-Lite$0.140/1M vs $0.250/1M input
- DeepSeek V4 Pro vs GPT-4o mini$0.435/1M vs $0.150/1M input
- o3 vs Claude Opus 4.6$2.00/1M vs $5.00/1M input
- o4-mini vs GPT-5 Mini$1.10/1M vs $0.250/1M input
- Grok 4 vs GPT-5$3.00/1M vs $1.25/1M input
- Mistral Large 3 vs GPT-4.1 Mini$0.500/1M vs $0.400/1M input
- GPT-5 Nano vs Gemini 2.0 Flash-Lite$0.050/1M vs $0.075/1M input
- Command R vs GPT-4o mini$0.150/1M vs $0.150/1M input
- Command R+ vs Claude Sonnet 4.6$2.50/1M vs $3.00/1M input
- Llama 3.3 70B (Groq) vs Llama 3.3 70B (Together)$0.590/1M vs $0.880/1M input
- Grok 4 Fast vs GPT-5 Mini$0.200/1M vs $0.250/1M input
- DeepSeek V3 vs DeepSeek V3 (Together)$0.270/1M vs $1.25/1M input
- DeepSeek V3 vs DeepSeek V3 (Fireworks)$0.270/1M vs $0.900/1M input
- Mixtral 8x7B (Groq) vs Llama 3.3 70B (Groq)$0.240/1M vs $0.590/1M input
- Mixtral 8x22B (Together) vs Mixtral 8x22B (Fireworks)$1.20/1M vs $0.900/1M input
- Llama 3.1 8B Instant (Groq) vs Llama 3.1 8B (Together)$0.050/1M vs $0.180/1M input
- Command R7B vs GPT-5 Nano$0.037/1M vs $0.050/1M input
- Claude 3.5 Haiku vs Claude Haiku 4.5$0.800/1M vs $1.00/1M input
- Gemini 1.5 Flash vs Gemini 2.5 Flash$0.075/1M vs $0.300/1M input
- o1 vs o3$15.00/1M vs $2.00/1M input
- Mistral Nemo vs Mistral Small$0.150/1M vs $0.060/1M input
- Llama 3.1 70B (Together) vs Llama 3.3 70B (Groq)$0.880/1M vs $0.590/1M input
- Llama 3.1 70B (Fireworks) vs Llama 3.1 70B (Together)$0.900/1M vs $0.880/1M input
- Claude 3 Opus vs Claude Opus 4.8$15.00/1M vs $5.00/1M input
- Claude 3 Sonnet vs Claude Sonnet 5$3.00/1M vs $2.00/1M input
- GPT-3.5 Turbo vs GPT-4o mini$0.500/1M vs $0.150/1M input
- DeepSeek Coder vs Codestral$0.140/1M vs $1.00/1M input
- Qwen2.5 Coder 32B (Together) vs Codestral$0.800/1M vs $1.00/1M input
- Llama 3.2 1B (Groq) vs Llama 3.2 3B (Groq)$0.040/1M vs $0.060/1M input
- Command Light vs Command R7B$0.300/1M vs $0.037/1M input
- GPT-4 vs GPT-4.1$30.00/1M vs $2.00/1M input
- Claude 3 Haiku vs Claude 3.5 Haiku$0.250/1M vs $0.800/1M input
- Gemini 1.5 Flash-8B vs Gemini 2.5 Flash-Lite$0.037/1M vs $0.100/1M input
- Open Mixtral 8x7B vs Mixtral 8x7B (Groq)$0.700/1M vs $0.240/1M input
- Llama 3.2 3B (Groq) vs Llama 3.2 3B (Together)$0.060/1M vs $0.060/1M input
- Llama 3.2 3B (Together) vs Llama 3.2 3B (Fireworks)$0.060/1M vs $0.100/1M input
- Gemma 7B (Groq) vs Gemma 2 9B (Groq)$0.070/1M vs $0.200/1M input
- Command R+ (08-2024) vs Command R+$2.50/1M vs $2.50/1M input
- DeepSeek VL vs Pixtral 12B$0.140/1M vs $0.150/1M input
- o1-mini vs o4-mini$1.10/1M vs $1.10/1M input
- Claude 3.5 Sonnet vs Claude Sonnet 5$3.00/1M vs $2.00/1M input
- DeepSeek R1 vs DeepSeek R1 (Together)$0.550/1M vs $3.00/1M input
- DeepSeek R1 vs DeepSeek R1 (Fireworks)$0.550/1M vs $0.900/1M input
- Llama 4 Maverick (Together) vs Llama 4 Maverick (Fireworks)$0.270/1M vs $0.220/1M input
- Llama 4 Scout (Groq) vs Llama 4 Maverick (Together)$0.110/1M vs $0.270/1M input
- Open Mixtral 8x22B vs Mixtral 8x22B (Together)$2.00/1M vs $1.20/1M input
- QwQ 32B (Groq) vs Qwen3 32B (Groq)$0.290/1M vs $0.290/1M input
- GPT-4.1 (2025-04-14) vs GPT-4.1$2.00/1M vs $2.00/1M input
- GPT-5 vs Claude Sonnet 5$1.25/1M vs $2.00/1M input
- GPT-4o vs Claude Sonnet 5$2.50/1M vs $2.00/1M input
- GPT-4o vs Gemini 2.5 Pro$2.50/1M vs $1.25/1M input
- GPT-5 Mini vs Gemini 2.5 Flash$0.250/1M vs $0.300/1M input
- GPT-4.1 Mini vs Gemini 2.5 Flash$0.400/1M vs $0.300/1M input
- Claude Sonnet 4.6 vs Gemini 2.5 Flash$3.00/1M vs $0.300/1M input
- Claude Opus 4.6 vs GPT-5$5.00/1M vs $1.25/1M input
- Gemini 3.1 Pro vs Claude Sonnet 5$2.00/1M vs $2.00/1M input
- GPT-4o mini vs Claude Haiku 4.5$0.150/1M vs $1.00/1M input
- GPT-5 Nano vs Claude Haiku 4.5$0.050/1M vs $1.00/1M input
- GPT-4.1 Nano vs DeepSeek V4 Flash$0.100/1M vs $0.140/1M input
- Ministral 8B vs GPT-5 Nano$0.150/1M vs $0.050/1M input
- Mistral Small vs GPT-4o mini$0.060/1M vs $0.150/1M input
- Command R7B vs Gemini 2.5 Flash-Lite$0.037/1M vs $0.100/1M input
- DeepSeek Chat vs GPT-4o mini$0.280/1M vs $0.150/1M input
- Grok 3 Mini vs GPT-5 Mini$0.300/1M vs $0.250/1M input
- o3 vs DeepSeek R1$2.00/1M vs $0.550/1M input
- o3-mini vs o4-mini$1.10/1M vs $1.10/1M input
- DeepSeek Reasoner vs o4-mini$0.550/1M vs $1.10/1M input
- DeepSeek R1 Distill Qwen 32B vs DeepSeek R1 Distill Llama 70B (Groq)$0.300/1M vs $0.750/1M input
- o1-preview vs o1$15.00/1M vs $15.00/1M input
- Llama 3.3 70B (Groq) vs Llama 3.3 70B (Fireworks)$0.590/1M vs $0.900/1M input
- Llama 3.3 70B (Together) vs Llama 3.3 70B (Fireworks)$0.880/1M vs $0.900/1M input
- DeepSeek V3 (Together) vs DeepSeek V3 (Fireworks)$1.25/1M vs $0.900/1M input
- Qwen2.5 72B (Together) vs Qwen2.5 72B (Fireworks)$1.20/1M vs $0.900/1M input
- Llama 3.1 8B Instant (Groq) vs Llama 3.1 8B (Fireworks)$0.050/1M vs $0.200/1M input
- Open Mixtral 8x22B vs Mixtral 8x22B (Fireworks)$2.00/1M vs $0.900/1M input
- DeepSeek R1 (Together) vs DeepSeek R1 (Fireworks)$3.00/1M vs $0.900/1M input
- Codestral vs GPT-4.1 Mini$1.00/1M vs $0.400/1M input
- DeepSeek Coder vs Qwen2.5 Coder 32B (Together)$0.140/1M vs $0.800/1M input
- Pixtral 12B vs Llama 3.2 11B Vision (Groq)$0.150/1M vs $0.180/1M input
- Grok 2 Vision vs Pixtral 12B$2.00/1M vs $0.150/1M input
- GPT-4 Turbo vs GPT-4o$10.00/1M vs $2.50/1M input
- GPT-4o vs GPT-4.1$2.50/1M vs $2.00/1M input
- Gemini 1.5 Pro vs Gemini 2.5 Pro$1.25/1M vs $1.25/1M input
- Gemini 2.0 Flash vs Gemini 2.5 Flash$0.100/1M vs $0.300/1M input
- Grok 3 vs Grok 4$3.00/1M vs $3.00/1M input
- Mistral Large (2407) vs Mistral Large 3$2.00/1M vs $0.500/1M input
- Command A vs Command R+$2.50/1M vs $2.50/1M input
- Claude 2.1 vs Claude Opus 4.8$8.00/1M vs $5.00/1M input
Related
Exact vs Approx · Prompt cost · Batch pricing · Cache savings · Pricing changelog