HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/zai-org/zai-org/GLM-4.7

zai-org/GLM-4.7

family glm · 7 variants · unknown

provider
zai-org
released
—
huggingface
https://huggingface.co/zai-org/GLM-4.7
best scores
Terminal-Bench: 33.3MMLU-Pro: 84.3GPQA Diamond: 85.7Humanity's Last Exam: 42.8AIME 2025: 95.7LiveCodeBench: 84.9SWE-bench Verified: 73.8Terminal-Bench 2.0: 41.0BrowseComp: 67.5

Best scores per benchmark

benchmarkcategorybest score
AIME 2025math95.7
GPQA Diamondreasoning85.7
LiveCodeBenchcoding84.9
MMLU-Proknowledge84.3
SWE-bench Verifiedswe73.8
BrowseCompagentic67.5
Humanity's Last Examknowledge42.8
Terminal-Bench 2.0agentic-coding41.0
Terminal-Benchagentic-coding33.3

Variants (7)

variantkindquantbitsvRAMdlorgyear
GLM-4.7originalbf1616.060,889zai-org2026
GLM-4.7-FP8fp8fp88.045,145zai-org2025
GLM-4.7-GGUFggufgguf16.06,594unsloth2025
GLM-4.7-NVFP4nvfp4nvfp44.04,945nvidia2026
zai-org_GLM-4.7-GGUFggufgguf1.01,954bartowski2025
GLM-4.7-GGUFggufgguf4.0215lmstudio-community2025
GLM-4.7-GGUFggufgguf16.079Mungert2026
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.