deepseek-ai/deepseek-r1 pricing
deepseek-ai/deepseek-r1, Reasoning-tier from Replicate, costs $3.75 per 1M input tokens and $10.00 per 1M output tokens.
Input$3.75per 1M tokens
Output$10.00per 1M tokens
Context window—tokens
TierReasoningReplicate
Price your own workload against deepseek-ai/deepseek-r1 in the calculator →
Specs and pricing
| Provider | Replicate |
| Tier | Reasoning |
| Context window | — |
| Max output | — |
| Input price | $3.75 / 1M tokens |
| Output price | $10.00 / 1M tokens |
| Modalities | text |
Note. Replicate bills output at $0.01 per thousand tokens ($10 per 1M, converted) and input at $3.75 per million tokens.
What it costs at realistic volumes
Assuming a 4:1 input:output token ratio, a common shape for chat and RAG-style workloads:
| Volume (4:1 in:out ratio) | Input cost | Output cost | Total / month |
|---|---|---|---|
| 1,000,000 in / 250,000 out | $3.75 | $2.50 | $6.25 |
| 10,000,000 in / 2,500,000 out | $37.50 | $25.00 | $62.50 |
| 100,000,000 in / 25,000,000 out | $375.00 | $250.00 | $625.00 |
Cheaper alternatives
Same tier (Reasoning), a different provider, and a lower blended cost at the 4:1 ratio used above:
| Model | Provider | $/1M in | $/1M out | At the 4:1 ratio |
|---|---|---|---|---|
| Trinity Large (Thinking) | Parasail | $0.22 | $0.85 | 93.1% cheaper blended |
| GPT-5.1-codex-mini | Azure OpenAI Service | $0.25 | $2.00 | 88% cheaper blended |
| QwQ 32B | Cloudflare Workers AI | $0.66 | $1.00 | 85.4% cheaper blended |
| DeepSeek-R1-Distill-Llama-70B | SambaNova | $0.70 | $1.40 | 83.2% cheaper blended |
| Kimi K2 Thinking | Novita AI | $0.60 | $2.50 | 80.4% cheaper blended |