Model

YandexGPT 5 Pro

CLOSED

Yandex · YandexGPT family · released Feb 25, 2025

Yandex's closed flagship LLM powering the Alice assistant; vendor claims quality comparable to GPT-4o, with an independent review putting MMLU around 83%.

ReasoningCodingVisionFunction callingTool useAgentic

976.2

Elo · rank #219

Parameters

Undisclosed

Active params

—

Context

32K tokens

Architecture

Proprietary dense transformer, undisclosed size (YandexGPT 5 flagship)

License

Proprietary

Languages

—

API price (in/out)

No hosted API

Modalities

text

Benchmark results

Bar shows position within the tracked field; marker = field best

BIG-Bench HardReasoning84.0%#20

best: ERNIE 4.5 300B-A47B · 94.3%

DROPReasoning63#49

best: Hunyuan-T1 · 93.1

GPQA DiamondReasoning44.9%#205

best: GPT-5.6 · 94.6%

HumanEvalCoding89.6%#20

best: Claude Opus 4.5 · 99.4%

IFEvalReasoning77.0%#101

best: Gemma 4 26B A4B · 98.5%

MATH-500Math57.5%#117

best: GPT-5 · 99.4%

MBPPCoding84.0%#8

best: Llama-3.3-Nemotron-Super-49B v1 (Reasoning On) · 91.3%

MMLU-ProKnowledge68.0%#98

best: Claude Fable 5 · 91.5%

MMLUKnowledge83.0%#75

best: OpenAI o3 · 92.9%

Run it locally

Closed weights — available via API only. No local deployment.

Input / M tok

—

Output / M tok

—

YandexGPT family

Elo progression across releases

YandexGPT 5 ProFeb 2025976.2

Resources

DOCSyandex.cloud/en/docs/foundation-models APIAvailable via Yandex API

API price — · each benchmark row carries its own source badge (see methodology)