Nanbeige4.1-3B

Nanbeige

Hugging Face ↗

Verified against Artificial Analysis · Jul 13, 2026

Parameters

Price /M

$0 / $0

in / out

Context

—

Avg score

25.5

7 benchmarks

AA Intelligence

11.1

index

Weights

Open

Released

Feb 11, 2026

About

Nanbeige4.1-3B is a 3B-parameter open-weights model in the Nanbeige4 family from Nanbeige. Benchmarked on 7 evals, averaging 25.5.

apache-2.0text-generation

Benchmark scores · 7

avg 25.5 — higher is better

GPQA Diamondartificial-analysis · 2026-07

84.9

IFBenchartificial-analysis · 2026-07

35.4

SciCodeartificial-analysis · 2026-07

26.6

τ²-benchartificial-analysis · 2026-07

21.6

Humanity's Last Examartificial-analysis · 2026-07

Terminal-Bench Hardartificial-analysis · 2026-07

AA Long-Context Reasoningartificial-analysis · 2026-07

The research behind this model

all papers mentioning it →

Nanbeige4.1-3B: A Small General Model that Reasons, Aligns, and Acts

Details

PublisherNanbeige

FamilyNanbeige4

Parameters3B

Licenseapache-2.0

WeightsOpen

ReleasedFeb 11, 2026

Performance & price

Price in / out$0 / $0 per 1M

Coding index9.6

Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →

History

full ledger →

Capability over time

Append-only ledger — every observed change to this model's numbers.

API providers

Artificial Analysis$0/M