All AI models · LLM benchmarks · Methodology

DeepSeek

DeepSeek V4 Pro

Open MoE flagship with million-token context for coding and long agent runs

Model facts

Context window
1M tokens
Maximum output
384K tokens
Cheapest paid input
$0.348 per 1M tokens
Cheapest paid output
$0.696 per 1M tokens
Open weights
yes
Providers
79

Capabilities and modalities

reasoning, tool calling, structured output, attachments, temperature control, open weights, text.

Local hardware estimate

At 4K context: Q4_K_M 979.5 GiB, Q8_0 1724.1 GiB, F16 3120.0 GiB working memory.

Parameters
1598.8 billion
Architecture
deepseekv4
Model license
mit

Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.

Artificial Analysis benchmarks

Compare this model on the full LLM benchmark leaderboard.

ConfigurationAAICodingMathOutput tokens/s
Reasoning, Max Effort45.359.4not measured67.8
Reasoning, High Effort43.758.7not measured66.5
Non-reasoning31.9not measurednot measured67.4

Observed price history

  • 2026-08-24: $0.348 input / $0.696 output per 1M tokens

API providers

  • Alibaba Token Plan — model id deepseek-v4-pro; free input / free output per 1M tokens; provider documentation
  • Alibaba Token Plan (China) — model id deepseek-v4-pro; free input / free output per 1M tokens; provider documentation
  • Kenari — model id deepseek-v4-pro; free input / free output per 1M tokens; provider documentation
  • SCNet Token Plan — model id DeepSeek-V4-Pro; free input / free output per 1M tokens; provider documentation
  • Umans AI Coding Plan — model id umans-deepseek-v4-pro-0813; free input / free output per 1M tokens; provider documentation
  • UnoRouter — model id deepseek-v4-pro:free; free input / free output per 1M tokens; provider documentation
  • Volcengine Ark Coding Plan — model id deepseek-v4-pro; free input / free output per 1M tokens; provider documentation
  • routing.run — model id deepseek-v4-pro; $0.348 input / $0.696 output per 1M tokens; provider documentation
  • CrofAI — model id deepseek-v4-pro; $0.35 input / $0.8 output per 1M tokens; provider documentation
  • EBCloud — model id DeepSeek-V4-Pro; $0.429 input / $0.857 output per 1M tokens; provider documentation
  • Alibaba (China) — model id deepseek-v4-pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • Auriko — model id deepseek-v4-pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • DeepSeek — model id deepseek-v4-pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • DevPass (LLM Gateway) — model id deepseek-v4-pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • Hugging Face — model id deepseek-ai/DeepSeek-V4-Pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • LLM Gateway (DeepSeek) — model id deepseek/deepseek-v4-pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • Modelis — model id deepseek-v4-pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • Nvidia — model id deepseek-ai/deepseek-v4-pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • Pioneer — model id deepseek-ai/DeepSeek-V4-Pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • TokenGo — model id deepseek/deepseek-v4-pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • Vivgrid — model id deepseek-v4-pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • ZenMux — model id deepseek/deepseek-v4-pro; $0.435 input / $0.87 output per 1M tokens; provider documentation
  • OrcaRouter — model id deepseek/deepseek-v4-pro; $0.442 input / $0.884 output per 1M tokens; provider documentation
  • AIHubMix (DeepSeek) — model id deep-deepseek-v4-pro; $0.478 input / $0.956 output per 1M tokens; provider documentation
  • DigitalOcean — model id deepseek-v4-pro; $0.87 input / $1.74 output per 1M tokens; provider documentation
  • Merge Gateway — model id deepseek/deepseek-v4-pro; $0.66 input / $1.98 output per 1M tokens; provider documentation
  • OpenCode Go (New) — model id deepseek-v4-pro; $0.66 input / $1.98 output per 1M tokens; provider documentation
  • Vancine — model id deepseek-v4-pro; $0.66 input / $1.98 output per 1M tokens; provider documentation
  • Vercel AI Gateway — model id deepseek/deepseek-v4-pro; $0.66 input / $1.98 output per 1M tokens; provider documentation
  • UnoRouter — model id deepseek-v4-pro; $0.9 input / $1.8 output per 1M tokens; provider documentation
  • LLM Gateway (Runware) — model id runware/deepseek-v4-pro; $0.961 input / $1.92 output per 1M tokens; provider documentation
  • above.dev — model id deepseek-v4-pro; $0.726 input / $2.18 output per 1M tokens; provider documentation
  • OpenRouter — model id deepseek/deepseek-v4-pro; $1.04 input / $2.08 output per 1M tokens; provider documentation
  • NanoGPT — model id deepseek/deepseek-v4-pro; $1.1 input / $2.2 output per 1M tokens; provider documentation
  • ai& — model id deepseek-ai/deepseek-v4-pro; $1 input / $2.5 output per 1M tokens; provider documentation
  • Weights & Biases — model id deepseek-ai/DeepSeek-V4-Pro; $1.15 input / $2.55 output per 1M tokens; provider documentation
  • Deep Infra — model id deepseek-ai/DeepSeek-V4-Pro; $1.3 input / $2.6 output per 1M tokens; provider documentation
  • LLM Gateway (DeepInfra) — model id deepinfra/deepseek-v4-pro; $1.3 input / $2.6 output per 1M tokens; provider documentation
  • Neuralwatt — model id deepseek-v4-pro; $1 input / $3 output per 1M tokens; provider documentation
  • GMI Cloud — model id deepseek-ai/DeepSeek-V4-Pro; $1.39 input / $2.78 output per 1M tokens; provider documentation
  • Kilo Gateway — model id deepseek/deepseek-v4-pro; $1.6 input / $3.2 output per 1M tokens; provider documentation
  • NovitaAI — model id deepseek/deepseek-v4-pro; $1.6 input / $3.2 output per 1M tokens; provider documentation
  • CrossModel — model id deepseek/deepseek-v4-pro; $1.22 input / $3.65 output per 1M tokens; provider documentation
  • EmpirioLabs AI — model id deepseek-v4-pro; $1.65 input / $3.3 output per 1M tokens; provider documentation
  • Venice AI — model id deepseek-v4-pro; $1.65 input / $3.3 output per 1M tokens; provider documentation
  • Jalapeno Cloud — model id DeepSeek-V4-Pro; $1.6 input / $3.38 output per 1M tokens; provider documentation
  • AIHubMix (Alibaba Cloud) — model id alicloud-deepseek-v4-pro; $1.69 input / $3.38 output per 1M tokens; provider documentation
  • Cortecs — model id deepseek-v4-pro; $1.73 input / $3.46 output per 1M tokens; provider documentation
  • Abacus — model id deepseek-ai/DeepSeek-V4-Pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • Arcee — model id deepseek/deepseek-v4-pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • Azure — model id deepseek-v4-pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • Baseten — model id deepseek-ai/DeepSeek-V4-Pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • ClinePass — model id cline-pass/deepseek-v4-pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • Cloudflare AI Gateway — model id deepseek/deepseek-v4-pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • FastRouter — model id deepseek/deepseek-v4-pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • FrogBot — model id deepseek-v4-pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • HPC-AI — model id deepseek/deepseek-v4-pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • Impossibl — model id deepseek/deepseek-v4-pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • LLM Gateway (CanopyWave) — model id canopywave/deepseek-v4-pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • SiliconFlow — model id deepseek-ai/DeepSeek-V4-Pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • Together AI — model id deepseek-ai/DeepSeek-V4-Pro; $1.74 input / $3.48 output per 1M tokens; provider documentation
  • Nebius Token Factory — model id deepseek-ai/DeepSeek-V4-Pro; $1.75 input / $3.5 output per 1M tokens; provider documentation
  • Requesty (EU) — model id deepseek-v4-pro@eu; $1.75 input / $3.5 output per 1M tokens; provider documentation
  • TensorX — model id deepseek/deepseek-v4-pro; $1.75 input / $3.5 output per 1M tokens; provider documentation
  • Eden AI — model id deepseek/deepseek-v4-pro; $1.32 input / $3.96 output per 1M tokens; provider documentation
  • LLM Gateway (Baidu) — model id baidu/deepseek-v4-pro; $1.32 input / $3.96 output per 1M tokens; provider documentation
  • LLM Gateway (ByteDance) — model id bytedance/deepseek-v4-pro; $1.32 input / $3.96 output per 1M tokens; provider documentation
  • LLM Gateway (Fireworks AI) — model id fireworks/deepseek-v4-pro; $1.32 input / $3.96 output per 1M tokens; provider documentation
  • LLM Gateway (Together AI) — model id together-ai/deepseek-v4-pro; $1.32 input / $3.96 output per 1M tokens; provider documentation
  • Ofox — model id deepseek/deepseek-v4-pro; $1.32 input / $3.96 output per 1M tokens; provider documentation
  • Requesty — model id deepseek-v4-pro; $1.32 input / $3.96 output per 1M tokens; provider documentation
  • Umans AI — model id umans-deepseek-v4-pro-0813; $1.32 input / $3.96 output per 1M tokens; provider documentation
  • OpenCode Zen — model id deepseek-v4-pro; $1.74 input / $3.84 output per 1M tokens; provider documentation
  • Charm Hyper — model id deepseek-v4-pro; $2.4 input / $4.8 output per 1M tokens; provider documentation
  • LLM Gateway (Alibaba Cloud) — model id alibaba/deepseek-v4-pro; $2.4 input / $4.8 output per 1M tokens; provider documentation
  • AnyAPI — model id deepseek/deepseek-v4-pro; not published input / not published output per 1M tokens; provider documentation
  • Arena — model id deepseek-v4-pro; not published input / not published output per 1M tokens; provider documentation
  • Model Oracle AI — model id deepseek-v4-pro; not published input / not published output per 1M tokens; provider documentation
  • Ollama Cloud — model id deepseek-v4-pro; not published input / not published output per 1M tokens; provider documentation