DeepSeek

DeepSeek R1 (0528)

164K ctxText Reasoning

Released May 28, 2025 · 148B tokens this week · 5 providers

Overview

The refreshed R1 reasoning model, which emits a full chain of thought before answering. Very strong on mathematics and algorithmic problems relative to its price. Expect high completion-token counts, since the thinking is billed as output.

reasoningacademiaprogramming

Providers

Reference data — not yet measured from live traffic

PROVIDER		MAX OUT		OUTPUT /M				QUANT
DeepSeekCheapest	164K	66K	$0.50	$2.15	104ms	46	99.30%	Full
Novita AI	164K	66K	$0.53	$2.26	780ms	70	98.50%	fp8
Together AI	164K	66K	$0.55	$2.37	195ms	82	99.45%	Full
Fireworks AI	164K	66K	$0.57	$2.45	660ms	96	99.50%	Full
Lambdadegraded	66K	66K	$0.59	$2.54	900ms	52	97.90%	fp8

Latency p50 reflects a provider’s API endpoint responsiveness — the round-trip to its API, not per-token inference time. These figures, with throughput and uptime, are reference data until the gateway aggregates its own traffic.

Price comparison

Blended $ per 1M tokens

Grok 4 Fast$0.42

DeepSeek R1 (0528)$1.74

Gemini 3 Flash$1.95

GPT-5.2$10.94

Claude Sonnet 4.5$12.00

Blended rate per 1M tokens, weighted one part prompt to three parts completion — roughly the shape of a chat workload. Your mix will move the number.

Call it

OpenAI-compatible — swap the base URL and go

curl https://model.cards/api/v1/chat/completions \
  -H "Authorization: Bearer $MODELCARDS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-r1-0528",
    "messages": [
      { "role": "user", "content": "Summarize the tradeoffs of speculative decoding." }
    ]
  }'