모델 목록으로
DeepSeek오픈소스

DeepSeek-V3.1

DeepSeek-V3.1, developed by DeepSeek-AI, is a large-scale foundation model with an approximate parameter scale of 6.71 trillion. It features a long 128K context window and provides advanced chat capabilities.

파라미터

6710.0B

컨텍스트

128K

라이선스

MIT

출시일

2025-08-20

API 가격

이 모델의 API 가격 정보는 현재 공개되지 않았습니다

강점

  • Extremely large parameters of 6710B
  • Long 128K context window
  • High openness under MIT license

약점

  • Massive model size of 1340GB
  • Requires very advanced computational resources to run
  • Difficult to operate on lightweight devices

활용 사례

  • Building advanced conversational AI chatbots
  • Long document analysis and processing
  • Customization of large-scale foundation models

심층 분석

Parameters

671B MoE

GPQA Diamond

80.1% (thinking)

AIME 2025

88.4% (thinking)

SWE-Bench Verified

66.0%

Pricing

$0.28/$0.42 per 1M tokens

Release Date

August 21, 2025

강점

  • Hybrid thinking/non-thinking mode
  • Strong agent and tool-use
  • Comparable to R1-0528 with faster responses
  • Extremely competitive pricing
  • MIT license

약점

  • Safety alignment concerns (Azure)
  • Non-thinking mode significantly weaker
  • SWE-bench trails frontier models
  • Superseded by V3.2

경쟁사 비교

ModelSWEGPQAPrice
Claude Sonnet 4.577.2%83.4%$3/$15
GPT-574.9%85.7%$2.50/$15
DeepSeek-V3.273.1%82.4%$0.28/$0.42

DeepSeek-V3.1 is a hybrid thinking/non-thinking MoE model. 671B params, strong agent capabilities, extremely competitive pricing.

분석 생성일: 2026-05-24