Compare

DeepSeek V4 vs DeepSeek R2

DeepSeek vs DeepSeek. Specs, benchmarks, and real per-task cost — all in one page.

Verdict

DeepSeek V4 is ~5.5× cheaper on a 3:1 input:output blend.

DeepSeek

The shock-the-market moment of 2026. Frontier coding quality at 1/50th the price of Opus.

Unbeatable $/qualityOpen weightsGreat at code

DeepSeek

DeepSeek’s reasoning variant. Competes with o4 on math at a fraction of the cost.

Frontier math at open-weights pricingLong CoT traces

Pricing

Input / 1M$0.14$0.55

Output / 1M$0.28$2.19

Context128K128K

Max output16K64K

LicenseOpen weightsOpen weights

Released2026-02-202026-01-15

Benchmarks

LMSYS ELO1334.0—

MMLU Pro85.487.8

HumanEval91.590.1

SWE-Bench54.1—

MATH88.794.1

GPQA—71.2

IFEval85.8—

Per-task cost

Summarize a 1-hour meeting transcript$0.0022$0.0093

Review a 500-line pull request$0.0017$0.0088

Answer a customer support ticket$0.0006$0.0034

Extract structured data from a resume$0.0010$0.0049

Debug a stack trace with context$0.0014$0.0077

Per-call cost using published token counts for each task. Real-world prompts vary.

Add up to four models, tweak tasks live.