⚡Under $0.50/hr🧠VRAM Estimator⚖Compare GPUs🎁Free LLM APIs🎯Model Index
ReasoningMoEContext: 125K

DeepSeek R1

Comprehensive deployment profile and benchmark telemetry for DeepSeek R1. Self-hosting memory footprints, verified token economics, and direct API endpoints.

Verified Engineering Benchmarks

SWE-bench
49.2%
LiveCodeBench
65.9%
MATH-500
97.3%
MMLU
90.8%

VRAM Requirements & Sizing

FP16 Weights
1340 GB
INT4 / GGUF (Quantized)
670 GB
Recommended GPU
8x H200
🔗 Quick Actions
🖥️ Compatible GPUs & Self-Host Pricing

Deterministic VRAM math from entity graph. Green = fits in single GPU.

GPUVRAMFP16FP8INT4Calculator Link
H100 SXM580 GBOOMOOMOOMPre-filled →
H200141 GBOOMOOMOOMPre-filled →
B200192 GBOOMOOMOOMPre-filled →
A100 80GB80 GBOOMOOMOOMPre-filled →
L40S48 GBOOMOOMOOMPre-filled →
RTX 409024 GBOOMOOMOOMPre-filled →

Active API Providers & Free Tiers

ProviderInput / 1MOutput / 1MSpeedFree Tier Status
DeepSeek$0.55$2.1930 tok/sPaid API Only
Together AI$0.80$2.4025 tok/sPaid API Only
Fireworks$0.90$2.5028 tok/sPaid API Only
OpenRouter$0.70$2.5022 tok/sPaid API Only
NVIDIA NIM$0.10$0.15500 tok/sPaid API Only
GitHub Models$0.15$0.2030 tok/s⚡ 15 RPM / 150 RPD
SambaNova$0.00$0.00350 tok/s⚡ Free Developer Tier

OpenAI SDK Drop-in

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.deepseek.com/v1",
  apiKey: process.env.DEEPSEEK_API_KEY,
});

const response = await client.chat.completions.create({
  model: "deepseek-r1",
  messages: [{ role: "user", content: "Hello" }],
});

Related Engineering Guides & Analysis

What should I do next?