Unverified50% confidenceFactExact time
NVIDIA L4 (g6/SageMaker) natively supports bf16, fp8, and int8, and int4 is widened to 16 bits; NVIDIA T4G (g5g) supports bf16 (emulated) and int8, does not support fp8, and widens int4 to 16 bits; NeuronCore-v2 (inf2/trn1) supports bf16 and int8, fp8 has hardware support but the compiler has yet to catch up, and int4 is widened to bf16 and is 27 times slower per call.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
Related Entities
Related Claims
UnverifiedWhen the hardware does not support native low-precision tensor cores, int4 weights are unpacked and expanded to bf16 or fp16 before matrix multiplication, causing the bandwidth and capacity advantages to be lost; T4G widening int4 to 16 bits and Neuron widening int4 to bf16 both fall under this mechanism.78% similarUnverifiedNVIDIA B200于2024年发布，FP16算力达到4.5 PFLOPS级别71% similarUnverified32GB of VRAM can run FP16 and FP8 original models directly, without needing to use GGUF quantized versions.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1011099API
curl https://kongchang.com/api/v1/knowledge/claims/1011099MCP
get_claim(id=1011099)