Model comparison

Claude Opus 4.5 vs DeepSeek-V3.1-Terminus

Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 42.6 on the Noometry Index. DeepSeek-V3.1-Terminus costs 21× less per token, which makes it the better buy when Claude Opus 4.5's lead doesn't matter for your workload.

Last verified October 10, 2026. 14 shared benchmarks.

Claude Opus 4.5 Anthropic

50.5

Rank #39 Confirmed

DeepSeek-V3.1-Terminus

DeepSeek

42.6

Rank #96 Confirmed

Summary

They share 14 benchmarks with published results for both. Claude Opus 4.5 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 reasoning, where Claude Opus 4.5 leads 42.5 to 25.7.
The biggest single-benchmark swing is Kagi LLM Benchmark: 80.2% for Claude Opus 4.5 and 57.4% for DeepSeek-V3.1-Terminus.
DeepSeek-V3.1-Terminus is cheaper at $0.30 / $1 per million input/output tokens, against $5 / $25 for Claude Opus 4.5.
Claude Opus 4.5 accepts more context: 200K tokens versus 164K.
DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.

Side by side

Claude Opus 4.5 and DeepSeek-V3.1-Terminus specifications
	Claude Opus 4.5	DeepSeek-V3.1-Terminus
Provider	Anthropic	DeepSeek
Noometry Index	50.5	42.6
Released	2025-11-01	2025-09-22
Weights	Proprietary	Open
Context window	200K	164K
Max output	64K	66K
Input $ / M tokens	$5	$0.30
Output $ / M tokens	$25	$1
Results tracked	69	16

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

Category by category

Coding Claude Opus 4.5 leads

Claude Opus 4.5: 55.4 (#21), DeepSeek-V3.1-Terminus: 41.5 (#114)

Coding benchmarks
Benchmark	Claude Opus 4.5	DeepSeek-V3.1-Terminus
LMArena Coding	1504	1426
ALE-Bench	1,025	745.17
SWE-bench Verified	76.7%	—
SWE-bench Verified (bash only)	76.8%	—
LMArena WebDev	1493	—
SWE-bench Multilingual	70.7%	—
SciCode	—	40.6%
GSO	26.5%	—
WeirdML	63.7%	—
AlgoTune	1.77	—

Agentic & Tool Use Not comparable

Claude Opus 4.5: 47.2 (#11), DeepSeek-V3.1-Terminus: —

Agentic & Tool Use benchmarks
Benchmark	Claude Opus 4.5	DeepSeek-V3.1-Terminus
Terminal-Bench	63.1%	—
Berkeley Function Calling Leaderboard	77.5%	—
GDPval	45.5%	—
Remote Labor Index	3.8%	—
τ²-bench Airline	84%	—
τ²-bench Banking	24.7%	—
τ²-bench Retail	79.6%	—
τ²-bench Telecom	92.3%	—
Cybench	82%	—
DeepResearch Bench	54.8%	—
OSWorld	66.3%	—
BALROG	43.5%	—
LMArena Search	1180	—
METR Time Horizons	75%	—
Vending-Bench 2	4,967	—

Reasoning Claude Opus 4.5 leads

Claude Opus 4.5: 42.5 (#43), DeepSeek-V3.1-Terminus: 25.7 (#126)

Reasoning benchmarks
Benchmark	Claude Opus 4.5	DeepSeek-V3.1-Terminus
Kagi LLM Benchmark	80.2%	57.4%
LMArena Hard Prompts	1476	1427
DTBench	89.9%	81.3%
LMCA	44.5%	28.6%
ARC-AGI-2	37.6%	—
SimpleBench	62%	—
NYT Connections (extended)	52.5%	—
ARC-AGI-1	80%	—
CritPt	—	1.7%
Chess Puzzles	12%	—
EnigmaEval	11.9%	—
EBR-Bench	14.3%	—
Mystery Game Puzzles	22%	—
Epoch Capabilities Index	150.1	—
ForecastBench	60.7	—

Math Claude Opus 4.5 leads

Claude Opus 4.5: 38.5 (#85), DeepSeek-V3.1-Terminus: 35.7 (#138)

Math benchmarks
Benchmark	Claude Opus 4.5	DeepSeek-V3.1-Terminus
LMArena Math	1463	1402
FrontierMath (Tiers 1-3)	34.4%	—
FrontierMath Tier 4	4.9%	—
OTIS Mock AIME 2024-2025	86.1%	—
ProofBench	36%	—
FrontierMath (Feb 2025 set)	20.7%	—
FrontierMath Tier 4 (v1)	4.2%	—

Knowledge Not comparable

Claude Opus 4.5: 56.4 (#38), DeepSeek-V3.1-Terminus: —

Knowledge benchmarks
Benchmark	Claude Opus 4.5	DeepSeek-V3.1-Terminus
GPQA Diamond	86%	—
Humanity's Last Exam	25.2%	—
SimpleQA Verified	45.7%	—
Vectara Hallucination Rate	10.9%	—
LMArena Expert	1488	—

Multimodal Not comparable

Claude Opus 4.5: 31.1 (#99), DeepSeek-V3.1-Terminus: —

Multimodal benchmarks
Benchmark	Claude Opus 4.5	DeepSeek-V3.1-Terminus
GeoBench	75%	—
VPCT	40%	—
Furniture Assembly	28.3%	—
LMArena Document	1462	—

Multilingual Claude Opus 4.5 leads

Claude Opus 4.5: 54.5 (#45), DeepSeek-V3.1-Terminus: 52.3 (#87)

Multilingual benchmarks
Benchmark	Claude Opus 4.5	DeepSeek-V3.1-Terminus
LMArena Non-English	1438	1407
LMArena Russian	1448	1434
LMArena Chinese	1470	—
LMArena French	1471	—
LMArena German	1449	—
LMArena Japanese	1415	—
LMArena Korean	1424	—
LMArena Spanish	1455	—

Instruction Following Claude Opus 4.5 leads

Claude Opus 4.5: 77.7 (#20), DeepSeek-V3.1-Terminus: 74.4 (#96)

Instruction Following benchmarks
Benchmark	Claude Opus 4.5	DeepSeek-V3.1-Terminus
LMArena Instruction Following	1478	1406

Long Context Claude Opus 4.5 leads

Claude Opus 4.5: 46.4 (#21), DeepSeek-V3.1-Terminus: 43.5 (#88)

Long Context benchmarks
Benchmark	Claude Opus 4.5	DeepSeek-V3.1-Terminus
LMArena Longer Query	1480	1426
CL-bench	21.1%	—

Writing & Preference Claude Opus 4.5 leads

Claude Opus 4.5: 68.2 (#28), DeepSeek-V3.1-Terminus: 61.3 (#87)

Writing & Preference benchmarks
Benchmark	Claude Opus 4.5	DeepSeek-V3.1-Terminus
LMArena Text	1451	1420
LMArena Creative Writing	1445	1404
LMArena Multi-Turn	1466	1411
EQ-Bench Creative Writing	1687	—

Frequently asked questions

Is Claude Opus 4.5 better than DeepSeek-V3.1-Terminus?

Which is cheaper, Claude Opus 4.5 or DeepSeek-V3.1-Terminus?

DeepSeek-V3.1-Terminus is cheaper. It lists at $0.30 per million input tokens and $1 per million output tokens; Claude Opus 4.5 lists at $5 and $25.

Is Claude Opus 4.5 or DeepSeek-V3.1-Terminus better for coding?

Claude Opus 4.5 scores higher on coding benchmarks: 55.4 versus 41.5 in the Noometry coding category.

Which has the bigger context window?

Claude Opus 4.5 does, with 200K tokens against 164K.

How many benchmarks do Claude Opus 4.5 and DeepSeek-V3.1-Terminus share?

14 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and DeepSeek-V3.1-Terminus has 16.