Expired50% confidenceFactExact time
Subsequent versions of LLaMA expanded the vocabulary from 32,000 to 128,000.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/20/2026
First Seen
Valid until: 10/18/2026(expired)
Sources
Related Claims
UnverifiedLLaMA系列最大的版本达到70B（700亿参数）61% similarPartially VerifiedTaking the 65B-parameter LLaMA model as an example, traditional full fine-tuning requires more than 780GB of VRAM, whereas QLoRA requires only approximately 48GB.57% similarUnverifiedFull fine-tuning of Llama-3 (8B) requires approximately 80 GB or more of VRAM, while Llama-3 (70B) requires more than 400 GB.57% similarUnverifiedMeta的Llama 3.1 70B参数模型以FP16加载约需140GB，405B参数模型以FP16加载约需810GB，INT4量化后约202GB55% similarUnverified15,000行代码换算成Token大约在150,000到300,000之间，取决于代码密度与注释比例55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/570280API
curl https://kongchang.com/api/v1/knowledge/claims/570280MCP
get_claim(id=570280)