deepseek

DeepSeek: DeepSeek V3.1

deepseek/deepseek-chat-v3.1

A text-model candidate for comparing multilingual instructions and structured responses. Use the same prompt and rubric for each language rather than treating an English result as a multilingual benchmark.

text

Try in Playground ↗

Input / 1M tokens$0.25

Output / 1M tokens$0.95

Context window163,840

Maximum output32,768

Provider-supported parameters

These are upstream capabilities. Jevrouter supports the subset described in the API contract; chat input is currently text only.

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_pstreamdeveloper_role

Capabilities are imported from OpenRouter’s catalog. This does not mean every model has been independently tested by Jevrouter.

Catalog updated: 2026-09-22T08:51:50.371Z · Price version: cmubxr2ec07uajn0zb771skjb:3

Provider pricing fields · USD per indicated unit

{
  "prompt": "0.00000025",
  "completion": "0.00000095",
  "input_cache_read": "0.00000013"
}

JEVROUTER INTEGRATION NOTES · REVIEWED 2026-09-22

What to check in your application

Reasoning-mode controls outside the documented Jevrouter contract are not forwarded.
If JSON is required, use a supported response_format and validate the resulting object in your application.

Start with the example below, then test representative inputs against your own acceptance criteria.

Make a request

curl --no-buffer https://api.jevrouter.io/v1/chat/completions \
  -H "Authorization: Bearer $JEVROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d @- <<'JSON'
{
  "model": "deepseek/deepseek-chat-v3.1",
  "messages": [
    {
      "role": "user",
      "content": "Return a JSON object with language and intent for: 请帮我更新发票地址。"
    }
  ],
  "max_tokens": 1024
}
JSON

Plan your integration

Connect with the OpenAI Python SDK →Estimate this model’s cost →Budgets, request status and safe retries →

Related models to evaluate

Use identical inputs and scoring rules. No ranking is implied by this selection.

Qwen: Qwen3 Coder 480B A35B DeepSeek: R1

Source: OpenRouter model catalog. Imported descriptions and capabilities are provider declarations, not independent benchmark results. Jevrouter currently uses OpenRouter as its managed upstream.