Model comparison

DeepSeek-V3.2-Speciale vs INTELLECT-1

DeepSeek-V3.2-Speciale has enough public results to be ranked (#162); INTELLECT-1 does not yet, so treat this comparison as directional.

Last verified October 11, 2026. 0 shared benchmarks.

DeepSeek-V3.2-Speciale

DeepSeek

39.7

Rank #162 Reported

INTELLECT-1

Hugging Face

—

Unranked

Summary

DeepSeek-V3.2-Speciale has downloadable open weights; the other is API-only.

Side by side

DeepSeek-V3.2-Speciale and INTELLECT-1 specifications
	DeepSeek-V3.2-Speciale	INTELLECT-1
Provider	DeepSeek	Hugging Face
Noometry Index	39.7	—
Released	2025-12-01	2024-11-29
Weights	Open	Proprietary
Context window	128K	—
Max output	128K	—
Input $ / M tokens	$0.58	—
Output $ / M tokens	$1.68	—
Results tracked	3	7

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

Category by category

Coding Not comparable

DeepSeek-V3.2-Speciale: 40.4 (#140), INTELLECT-1: —

Coding benchmarks
Benchmark	DeepSeek-V3.2-Speciale	INTELLECT-1
WeirdML	46.7%	—

Reasoning Not comparable

DeepSeek-V3.2-Speciale: 32.9 (#73), INTELLECT-1: —

Reasoning benchmarks
Benchmark	DeepSeek-V3.2-Speciale	INTELLECT-1
SimpleBench	52.6%	—
BIG-Bench Hard	—	34.9%
Epoch Capabilities Index	—	100.86
HellaSwag	—	71.4%
WinoGrande	—	65.8%

Math Not comparable

DeepSeek-V3.2-Speciale: —, INTELLECT-1: —

Math benchmarks
Benchmark	DeepSeek-V3.2-Speciale	INTELLECT-1
GSM8K	—	38.6%

Knowledge Not comparable

DeepSeek-V3.2-Speciale: —, INTELLECT-1: —

Knowledge benchmarks
Benchmark	DeepSeek-V3.2-Speciale	INTELLECT-1
ARC (AI2) Challenge	—	54.5%
MMLU	—	49.9%

Writing & Preference Not comparable

DeepSeek-V3.2-Speciale: 46.0 (#222), INTELLECT-1: —

Writing & Preference benchmarks
Benchmark	DeepSeek-V3.2-Speciale	INTELLECT-1
EQ-Bench Creative Writing	1276	—

Frequently asked questions

Is DeepSeek-V3.2-Speciale better than INTELLECT-1?