Compare

DeepSeek V4 vs Claude Sonnet 4.6

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

Verdict

Claude Sonnet 4.6 leads on LMSYS ELO (1368 vs 1334). DeepSeek V4 is ~34.3× cheaper on a 3:1 input:output blend. DeepSeek V4 is open-weights, the other is proprietary.

DeepSeek

DeepSeek V4

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

Unbeatable $/qualityOpen weightsGreat at code

Anthropic

Claude Sonnet 4.6

The workhorse. Near-Opus quality at 1/5 the cost. The default choice for production code and agent workloads.

Excellent cost/quality ratioFast inferenceGreat for Cursor/Claude Code

Pricing

Input / 1M$0.14$3.00

Output / 1M$0.28$15.00

Context128K500K

Max output16K64K

LicenseOpen weightsProprietary

Released2026-02-202025-11-04

Benchmarks

LMSYS ELO1334.01368.0

MMLU Pro85.488.7

HumanEval91.592.1

SWE-Bench54.164.5

MATH88.785.0

GPQA—59.4

MMMU—74.1

IFEval85.890.5

Per-task cost

Summarize a 1-hour meeting transcript$0.0022$0.052

Review a 500-line pull request$0.0017$0.054

Answer a customer support ticket$0.0006$0.021

Extract structured data from a resume$0.0010$0.030

Debug a stack trace with context$0.0014$0.048

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

Tool

Interactive compare →

Add up to four models, tweak tasks live.