Noometry

Model comparison

Claude Opus 4.8 vs DeepSeek-R1

Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 42.3 on the Noometry Index. DeepSeek-R1 costs 11× less per token, which makes it the better buy when Claude Opus 4.8's lead doesn't matter for your workload.

Last verified . 31 shared benchmarks.

Claude Opus 4.8 Anthropic

60.7

Rank #13 Confirmed

DeepSeek-R1 DeepSeek

42.3

Rank #115 Confirmed

Summary

  • They share 31 benchmarks with published results for both. Claude Opus 4.8 scores higher in 9 categories and DeepSeek-R1 in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Claude Opus 4.8 leads 64.7 to 18.6.
  • The biggest single-benchmark swing is ARC-AGI-1: 92.5% for Claude Opus 4.8 and 21.2% for DeepSeek-R1.
  • DeepSeek-R1 is cheaper at $0.50 / $2.15 per million input/output tokens, against $5 / $25 for Claude Opus 4.8.
  • Claude Opus 4.8 accepts more context: 1M tokens versus 164K.

Side by side

Claude Opus 4.8DeepSeek-R1
ProviderAnthropicDeepSeek
Noometry Index60.742.3
Released2026-05-282025-01-20
WeightsProprietaryProprietary
Context window1M164K
Max output128K64K
Input $ / M tokens$5$0.50
Output $ / M tokens$25$2.15
Results tracked6552

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Claude Opus 4.8 leads

Claude Opus 4.8: 59.9 (#12), DeepSeek-R1: 46.3 (#68)

BenchmarkClaude Opus 4.8DeepSeek-R1
SciCode53.5%35.7%
WeirdML82.9%41.6%
LMArena Coding14901427
ALE-Bench1,564804.12
DeepSWE59%—
FrontierCode46.5%—
Aider Polyglot—71.4%
LMArena WebDev1556—
GSO47.1%—
LiveBench Coding—66.7%
AlgoTune—1.7

Agentic & Tool Use Claude Opus 4.8 leads

Claude Opus 4.8: 47.6 (#11), DeepSeek-R1: 30.7 (#75)

BenchmarkClaude Opus 4.8DeepSeek-R1
DeepResearch Bench50.2%35.1%
APEX-Agents48.9%—
OSWorld 2.020.6%—
Remote Labor Index8.3%—
τ²-bench Banking39.7%—
PostTrainBench33.8%—
BALROG—34.9%
GBAEval70.9%—
GDP.pdf24%—
LMArena Search1204—
METR Time Horizons—53.8%
Vending-Bench 25,787—

Reasoning Claude Opus 4.8 leads

Claude Opus 4.8: 64.7 (#16), DeepSeek-R1: 18.6 (#278)

BenchmarkClaude Opus 4.8DeepSeek-R1
ARC-AGI-272.1%1.3%
SimpleBench64.8%40.8%
Kagi LLM Benchmark88.8%69.4%
ARC-AGI-192.5%21.2%
CritPt20.9%1.1%
LMArena Hard Prompts14821416
Epoch Capabilities Index158.21141.29
ForecastBench59.960
NYT Connections (extended)91.1%—
Chess Puzzles34%—
EnigmaEval23.5%—
EBR-Bench28.6%—
LiveBench Reasoning—83.2%
Mystery Game Puzzles36%—
DTBench94.9%—
LiveBench Data Analysis—69.8%
LMCA57.5%—
Surface Evolver Bench87.5%—
Bench to the Future 30.14—
LiveBench—71.6%

Math Claude Opus 4.8 leads

Claude Opus 4.8: 78.4 (#13), DeepSeek-R1: 43.8 (#79)

BenchmarkClaude Opus 4.8DeepSeek-R1
OTIS Mock AIME 2024-202598.3%66.4%
LMArena Math14871400
FrontierMath (Tiers 1-3)80%—
FrontierMath Tier 456.1%—
MathArena Final-Answer Competitions91.8%—
ProofBench69%—
Omni-MATH—42.4%
LiveBench Math—80.7%
MATH Level 5—96.6%
FrontierMath (Feb 2025 set)47.2%—
FrontierMath Tier 4 (v1)31.3%—

Knowledge Claude Opus 4.8 leads

Claude Opus 4.8: 61.3 (#29), DeepSeek-R1: 44.5 (#87)

BenchmarkClaude Opus 4.8DeepSeek-R1
GPQA Diamond91%76.3%
LMArena Expert15021394
SimpleQA Verified53%—
MMLU-Pro—79.3%
Confabulations—12.7%
Vectara Hallucination Rate—11.3%
GPQA (HELM)—66.6%

Multimodal Not comparable

Claude Opus 4.8: 42.9 (#26), DeepSeek-R1: —

BenchmarkClaude Opus 4.8DeepSeek-R1
LMArena Vision1294—
Blueprint-Bench 214.5%—
Furniture Assembly42.5%—
LMArena Document1475—

Multilingual Claude Opus 4.8 leads

Claude Opus 4.8: 55.2 (#33), DeepSeek-R1: 52.4 (#85)

BenchmarkClaude Opus 4.8DeepSeek-R1
LMArena Non-English14501412
LMArena Chinese15071442
LMArena French14811417
LMArena German14721404
LMArena Japanese14401391
LMArena Korean14321360
LMArena Russian14741423
LMArena Spanish14661411

Instruction Following Claude Opus 4.8 leads

Claude Opus 4.8: 77.4 (#24), DeepSeek-R1: 72.0 (#143)

BenchmarkClaude Opus 4.8DeepSeek-R1
LMArena Instruction Following14761382
LiveBench Instruction Following—80.5%
IFEval—78.4%

Long Context Too close to call

Claude Opus 4.8: 45.4 (#35), DeepSeek-R1: 45.4 (#36)

BenchmarkClaude Opus 4.8DeepSeek-R1
LMArena Longer Query14831391
Fiction.LiveBench—75%

Writing & Preference Claude Opus 4.8 leads

Claude Opus 4.8: 72.0 (#16), DeepSeek-R1: 61.4 (#88)

BenchmarkClaude Opus 4.8DeepSeek-R1
LMArena Text14611428
LMArena Creative Writing14541405
EQ-Bench Creative Writing18401500
LMArena Multi-Turn14761405
Short-Story Creative Writing—83%
WildBench—82.8%
EQ-Bench 41281—
LiveBench Language—48.5%

Frequently asked questions

Is Claude Opus 4.8 better than DeepSeek-R1?

Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 42.3 on the Noometry Index. DeepSeek-R1 costs 11× less per token, which makes it the better buy when Claude Opus 4.8's lead doesn't matter for your workload.

Which is cheaper, Claude Opus 4.8 or DeepSeek-R1?

DeepSeek-R1 is cheaper. It lists at $0.50 per million input tokens and $2.15 per million output tokens; Claude Opus 4.8 lists at $5 and $25.

Is Claude Opus 4.8 or DeepSeek-R1 better for coding?

Claude Opus 4.8 scores higher on coding benchmarks: 59.9 versus 46.3 in the Noometry coding category.

Which has the bigger context window?

Claude Opus 4.8 does, with 1M tokens against 164K.

How many benchmarks do Claude Opus 4.8 and DeepSeek-R1 share?

31 benchmarks have published results for both models. Claude Opus 4.8 has 65 scored results on Noometry and DeepSeek-R1 has 52.

Related comparisons

Go deeper