Loading NexProTools...
Preparing interactive calculation engine
Preparing interactive calculation engine
Compare 2026 pricing, prompt caching economics, batch API discounts, and end-to-end latency waterfall across OpenAI, Anthropic, Google, Meta, and DeepSeek.
| Model & Provider | Context Window | Input / Output Rate | Cost / 1k Req | Est. Monthly Bill | Throughput & Latency |
|---|---|---|---|---|---|
GPT-4o OpenAI | 128k | In: $2.50/M Out: $10.00/M | $8.000 | $800.00 In: $300.0 | Out: $500.0 | 45 tps 11.46s total (TTFT: 350ms) |
GPT-4o-mini OpenAI | 128k | In: $0.15/M Out: $0.60/M | $0.480 | $48.00 In: $18.0 | Out: $30.0 | 110 tps 4.73s total (TTFT: 180ms) |
Claude 3.5 Sonnet Anthropic | 200k | In: $3.00/M Out: $15.00/M | $10.380 | $1,038.00 In: $288.0 | Out: $750.0 | 70 tps 7.46s total (TTFT: 320ms) |
Claude 3.5 Haiku Anthropic | 200k | In: $0.80/M Out: $4.00/M | $2.768 | $276.80 In: $76.8 | Out: $200.0 | 130 tps 4.01s total (TTFT: 160ms) |
Gemini 1.5 Pro Google Cloud | 2M | In: $1.25/M Out: $5.00/M | $3.813 | $381.25 In: $131.3 | Out: $250.0 | 85 tps 6.16s total (TTFT: 280ms) |
Gemini 1.5 Flash Google Cloud | 1M | In: $0.07/M Out: $0.30/M | $0.229 | $22.88 In: $7.9 | Out: $15.0 | 160 tps 3.25s total (TTFT: 120ms) |
Llama 3.3 70B (Groq) Meta / Groq | 128k | In: $0.59/M Out: $0.79/M | $1.280 | $128.00 In: $88.5 | Out: $39.5 | 280 tps 1.88s total (TTFT: 90ms) |
DeepSeek V3 / R1 DeepSeek | 64k | In: $0.14/M Out: $0.28/M | $0.274 | $27.44 In: $13.4 | Out: $14.0 | 65 tps 8.09s total (TTFT: 400ms) |