HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/tencent/tencent/Youtu-LLM-2B-Base

tencent/Youtu-LLM-2B-Base

family misc · 1 variants · unknown

provider
tencent
released
—
huggingface
https://huggingface.co/tencent/Youtu-LLM-2B-Base
best scores
MMLU: 53.0MMLU-Pro: 36.5IFEval: 29.4DROP: 41.3MUSR: 43.8MATH-500: 84.8GPQA Diamond: 33.6BBH: 31.0HumanEval: 64.0MBPP: 51.5LiveCodeBench: 19.8SWE-bench Verified: 60.0BFCL (function calling): 55.5

Best scores per benchmark

benchmarkcategorybest score
MATH-500math84.8
HumanEvalcoding64.0
SWE-bench Verifiedswe60.0
BFCL (function calling)agentic55.5
MMLUknowledge53.0
MBPPcoding51.5
MUSRreasoning43.8
DROPreasoning41.3
MMLU-Proknowledge36.5
GPQA Diamondreasoning33.6
BBHreasoning31.0
IFEvalinstruction-following29.4
LiveCodeBenchcoding19.8

Variants (1)

variantkindquantbitsvRAMdlorgyear
Youtu-LLM-2B-GGUFggufgguf16.0176Mungert2026
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.