HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/deepseek-ai/deepseek-ai/DeepSeek-V3.1-Terminus

deepseek-ai/DeepSeek-V3.1-Terminus

family deepseek · 2 variants · unknown

provider
deepseek-ai
released
—
huggingface
https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus
best scores
MMLU-Pro: 84.8GPQA Diamond: 80.1Humanity's Last Exam: 15.9LiveCodeBench: 74.8BrowseComp: 49.2SimpleQA: 93.4Terminal-Bench: 31.3

Best scores per benchmark

benchmarkcategorybest score
SimpleQAknowledge93.4
MMLU-Proknowledge84.8
GPQA Diamondreasoning80.1
LiveCodeBenchcoding74.8
BrowseCompagentic49.2
Terminal-Benchagentic-coding31.3
Humanity's Last Examknowledge15.9

Variants (2)

variantkindquantbitsvRAMdlorgyear
DeepSeek-V3.1-Terminus-GGUFggufgguf16.03,598unsloth2025
deepseek-ai_DeepSeek-V3.1-Terminus-GGUFggufgguf1.0561bartowski2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.