DeepSeek R1 vs Claude Opus 4.8: which is better?

Claude Opus 4.8 leads on combined benchmark scores (89.3 vs 88.2) across general knowledge, coding, and math, as of 2026-07-27.

Which is cheaper, DeepSeek R1 or Claude Opus 4.8?

DeepSeek R1 is cheaper on a typical API workload, at roughly 10.3x lower cost than the alternative at published per-token rates.

DeepSeek R1 vs Claude Opus 4.8

Claude Opus 4.8 leads on benchmarks; DeepSeek R1 is cheaper to run, about 10.3× less.

Updated 2026-07-27

DeepSeek R1

DeepSeek · Open

88.2

Combined index

Claude Opus 4.8

Anthropic · Closed

89.3

Combined index

Leads

Benchmarks

General knowledge (MMLU)

86.9

89.5

Coding

82.5

Math

95.1

92.4

API pricing

$0.55 Input / 1M $5.0

$2.19 Output / 1M $25.0

$21.95 Monthly, est. $225.0

Monthly estimate assumes 20M input + 5M output tokens. Prices marked ~ are representative hosted rates for open-weight models and vary by provider. Run your own numbers in the cost calculator.

DeepSeek R1 costs → Claude Opus 4.8 costs → Full leaderboard → All comparisons →