Noometry

Model comparison

Claude Opus 4.8 vs DeepSeek-V3.1-Terminus

Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 43.1 on the Noometry Index. DeepSeek-V3.1-Terminus costs 22× less per token, which makes it the better buy when Claude Opus 4.8's lead doesn't matter for your workload.

Last verified . 16 shared benchmarks.

Claude Opus 4.8 Anthropic

60.7

Rank #13 Confirmed

DeepSeek-V3.1-Terminus DeepSeek

43.1

Rank #97 Confirmed

Summary

  • They share 16 benchmarks with published results for both. Claude Opus 4.8 scores higher in 7 categories and DeepSeek-V3.1-Terminus in 0 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Opus 4.8 leads 78.4 to 38.5.
  • The biggest single-benchmark swing is Kagi LLM Benchmark: 88.8% for Claude Opus 4.8 and 57.4% for DeepSeek-V3.1-Terminus.
  • DeepSeek-V3.1-Terminus is cheaper at $0.27 / $1 per million input/output tokens, against $5 / $25 for Claude Opus 4.8.
  • Claude Opus 4.8 accepts more context: 1M tokens versus 164K.
  • DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.

Side by side

Claude Opus 4.8DeepSeek-V3.1-Terminus
ProviderAnthropicDeepSeek
Noometry Index60.743.1
Released2026-05-282025-09-22
WeightsProprietaryOpen
Context window1M164K
Max output128K147K
Input $ / M tokens$5$0.27
Output $ / M tokens$25$1
Results tracked6516

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-V3.1-Terminus: 42.0 (#113)

BenchmarkClaude Opus 4.8DeepSeek-V3.1-Terminus
SciCode53.5%40.6%
LMArena Coding14901426
ALE-Bench1,564745.17
DeepSWE59%—
FrontierCode46.5%—
LMArena WebDev1556—
GSO47.1%—
WeirdML82.9%—

Agentic & Tool Use Not comparable

Claude Opus 4.8: 47.6 (#11), DeepSeek-V3.1-Terminus: —

BenchmarkClaude Opus 4.8DeepSeek-V3.1-Terminus
APEX-Agents48.9%—
OSWorld 2.020.6%—
Remote Labor Index8.3%—
τ²-bench Banking39.7%—
DeepResearch Bench50.2%—
PostTrainBench33.8%—
GBAEval70.9%—
GDP.pdf24%—
LMArena Search1204—
Vending-Bench 25,787—

Reasoning Claude Opus 4.8 leads

Claude Opus 4.8: 64.7 (#16), DeepSeek-V3.1-Terminus: 26.4 (#133)

BenchmarkClaude Opus 4.8DeepSeek-V3.1-Terminus
Kagi LLM Benchmark88.8%57.4%
CritPt20.9%1.7%
LMArena Hard Prompts14821426
DTBench94.9%81.3%
LMCA57.5%28.6%
ARC-AGI-272.1%—
SimpleBench64.8%—
NYT Connections (extended)91.1%—
ARC-AGI-192.5%—
Chess Puzzles34%—
EnigmaEval23.5%—
EBR-Bench28.6%—
Mystery Game Puzzles36%—
Surface Evolver Bench87.5%—
Bench to the Future 30.14—
Epoch Capabilities Index158.21—
ForecastBench59.9—

Math Claude Opus 4.8 leads

Claude Opus 4.8: 78.4 (#13), DeepSeek-V3.1-Terminus: 38.5 (#137)

BenchmarkClaude Opus 4.8DeepSeek-V3.1-Terminus
LMArena Math14871402
FrontierMath (Tiers 1-3)80%—
FrontierMath Tier 456.1%—
MathArena Final-Answer Competitions91.8%—
OTIS Mock AIME 2024-202598.3%—
ProofBench69%—
FrontierMath (Feb 2025 set)47.2%—
FrontierMath Tier 4 (v1)31.3%—

Knowledge Not comparable

Claude Opus 4.8: 61.3 (#29), DeepSeek-V3.1-Terminus: —

BenchmarkClaude Opus 4.8DeepSeek-V3.1-Terminus
GPQA Diamond91%—
SimpleQA Verified53%—
LMArena Expert1502—

Multimodal Not comparable

Claude Opus 4.8: 42.9 (#26), DeepSeek-V3.1-Terminus: —

BenchmarkClaude Opus 4.8DeepSeek-V3.1-Terminus
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-V3.1-Terminus: 52.1 (#92)

BenchmarkClaude Opus 4.8DeepSeek-V3.1-Terminus
LMArena Non-English14501407
LMArena Russian14741436
LMArena Chinese1507—
LMArena French1481—
LMArena German1472—
LMArena Japanese1440—
LMArena Korean1432—
LMArena Spanish1466—

Instruction Following Claude Opus 4.8 leads

Claude Opus 4.8: 77.4 (#24), DeepSeek-V3.1-Terminus: 74.0 (#106)

BenchmarkClaude Opus 4.8DeepSeek-V3.1-Terminus
LMArena Instruction Following14761404

Long Context Claude Opus 4.8 leads

Claude Opus 4.8: 45.4 (#35), DeepSeek-V3.1-Terminus: 43.4 (#97)

BenchmarkClaude Opus 4.8DeepSeek-V3.1-Terminus
LMArena Longer Query14831421

Writing & Preference Claude Opus 4.8 leads

Claude Opus 4.8: 72.0 (#16), DeepSeek-V3.1-Terminus: 61.0 (#92)

BenchmarkClaude Opus 4.8DeepSeek-V3.1-Terminus
LMArena Text14611419
LMArena Creative Writing14541403
LMArena Multi-Turn14761411
EQ-Bench Creative Writing1840—
EQ-Bench 41281—

Frequently asked questions

Is Claude Opus 4.8 better than DeepSeek-V3.1-Terminus?

Claude Opus 4.8 is the stronger model overall, scoring 60.7 to 43.1 on the Noometry Index. DeepSeek-V3.1-Terminus costs 22× 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-V3.1-Terminus?

DeepSeek-V3.1-Terminus is cheaper. It lists at $0.27 per million input tokens and $1 per million output tokens; Claude Opus 4.8 lists at $5 and $25.

Is Claude Opus 4.8 or DeepSeek-V3.1-Terminus better for coding?

Claude Opus 4.8 scores higher on coding benchmarks: 59.9 versus 42.0 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-V3.1-Terminus share?

16 benchmarks have published results for both models. Claude Opus 4.8 has 65 scored results on Noometry and DeepSeek-V3.1-Terminus has 16.

Related comparisons

Go deeper