anoman

DeepSeek

DeepSeek R1

modeldeepseek-reasoner
BudgetStableCommunityStreamingReasoningLong contextChina

Context

131K

Input / 1M tokens

Rp 5.175

Output / 1M tokens

Rp 7.762

Cached input / 1M tokens

Rp 1.216

Cache write / 1M tokens

โ€” (bills at input price)

About this model

DeepSeek R1

DeepSeek R1 via direct DeepSeek endpoint. Same weights as our `deepseek-r1` slug.

Key facts

Maker
DeepSeek
Context window
131K tokens
Max output
66K tokens
Lane
Community: needs account opt-in; the provider may use data for training
Modality
Text
Input price
Rp 5.175 per 1M tokens
Output price
Rp 7.762 per 1M tokens
Processing region(s)
cn
Usable on
Free (from credit or balance), Plus Pass, Pro, Enterprise

More models by DeepSeek

Released

2025-01

Training cutoff

2024-07

Parameters

MoE 671B (37B active)

Best for

Use cases

  • Hard reasoning + math
  • Same use cases as DeepSeek R1

Strengths

What it does well

  • Direct provider route
  • Identical weights to deepseek-r1

Limitations

Know the trade-offs

  • Same model, higher cost than alternative route

Code example

Chat completion with DeepSeek R1

from openai import OpenAI

client = OpenAI(
    base_url="https://api.anoman.io/v1",
    api_key="anm-sk-..."
)

response = client.chat.completions.create(
    model="deepseek-reasoner",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)

Use DeepSeek R1 through Anoman.