Last refreshed 2026-05-22. Next refresh: weekly.
Why use DeepSeek V3.1 on Replicate API?
Replicate API offers DeepSeek V3.1 with pay-as-you-go pricing at $0.67/1M input tokens. Replicate is a cloud-based platform that enables users to run machine learning models easily and efficiently.
Compare DeepSeek V3.1 across 9 providers to find the best fit for your use caseInput / 1M
$0.672
Output / 1M
$2.02
Cache
Not sourced
Batch
Not sourced
Setup recipe
Python + curlInstall
pip install replicateAuth
export REPLICATE_API_TOKEN=...Call
import replicate
output = replicate.run(
"deepseek-ai/deepseek-v3.1",
input={"prompt": "Hello"}Model ID
deepseek-ai/deepseek-v3.1Request example
import replicate
# reads REPLICATE_API_TOKEN from env
# deepseek-ai/deepseek-v3.1 format: "owner/model-name" (latest version) or "owner/model-name:version-hash"
output = replicate.run(
"deepseek-ai/deepseek-v3.1",
input={"prompt": "Hello"}
)
# Output is a list or generator depending on the model
print("".join(output))Gotchas
- Use provider model ID "deepseek-ai/deepseek-v3.1", not the LLMReference slug "deepseek-v3.1".
- Replicate uses "owner/model-name" format (e.g. "meta/meta-llama-3-8b-instruct") for the latest version, or "owner/model-name:version-sha" to pin to a specific version. The REST endpoint splits owner and model-name into the path: /v1/models/{owner}/{model-name}/predictions.
- The examples expect REPLICATE_API_TOKEN; rename it only if your application config maps the new variable.
Compare DeepSeek V3.1 Across Providers
| Provider | Input (per 1M) | Output (per 1M) |
|---|---|---|
| Microsoft Foundry | — | — |
| Fireworks AI | $0.56 | $1.68 |
| NVIDIA NIM | — | — |
| Together AI | $0.60 | $1.70 |
| AWS Bedrock | $0.60 | $1.73 |
Pricing
| Type | Price (per 1M) |
|---|---|
| Input tokens | $0.67 |
| Output tokens | $2.02 |
Capabilities
VisionMultimodalStructured OutputsCode Execution
About DeepSeek V3.1
Enhanced reasoning and grounded retrieval model from DeepSeek with multimodal text and image understanding.
Get Started
Model Specs
Released2025-08-21
Parameters671B total, 37B active (MoE)
Context64k
ArchitectureMixture of Experts