Expired50% confidenceFactExact time
OpenAI's text-embedding-3-small outputs 1536-dimensional vectors, while BGE-large-zh outputs 1024-dimensional vectors.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Related Claims
UnverifiedOpenAI's text-embedding-3 or the open-source BGE-M3 can generate dense vectors of 1536 dimensions or more81% similarUnverifiedOfficial reports indicate that trimming text-embedding-3-large from 3072 dimensions to 256 dimensions results in only approximately 2–3% loss in retrieval accuracy on the MTEB benchmark.63% similarUnverifiedLlama 3 8B has 32 layers, 8 KV heads (GQA architecture with 32 query heads), a head dimension of 128, and a KV cache of approximately 1.07 GB for a single sequence under FP16 with an 8192 context length.59% similarUnverifiedAt the maximum size and lowest error correction level, a single QR code can carry approximately 2,953 bytes of binary data.58% similarUnverifiedUnsloth supports quantized export in GGUF format, making it possible to fine-tune models with 7B or even 13B parameters on a single RTX 4090 with 24GB of VRAM58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/40811API
curl https://kongchang.com/api/v1/knowledge/claims/40811MCP
get_claim(id=40811)