BitStack
LLM quantization
present, but with little supersession signal in the knowledge base
0 papers critique it · 1 beat it on benchmarks
Beaten on benchmarks
Head-to-head results where a newer method reports beating BitStack. Values are copied from the source paper's tables — verify against the cited paper.
- AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models
AMQ beats BitStack · Avg. [Llama 2 7B, 2.5-bit]
60.56 vs 59.91
- AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models
AMQ beats BitStack · Avg. [Llama 2 7B, 3.0-bit]
65.59 vs 63.63
- AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models
AMQ beats BitStack · Avg. [Llama 2 7B, 3.5-bit]
68.41 vs 65.54
- AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models
AMQ beats BitStack · Avg. [Llama 2 13B, 2.5-bit]
67.05 vs 65.21
- AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models
AMQ beats BitStack · Avg. [Llama 2 13B, 3.0-bit]
70.31 vs 69.25
- AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models
AMQ beats BitStack · Avg. [Llama 2 13B, 3.5-bit]
72.34 vs 70.87
- AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models
AMQ beats BitStack · Avg. [Llama 2 70B, 2.5-bit]
73.75 vs 72.92
- AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models
AMQ beats BitStack · Avg. [Llama 2 70B, 3.0-bit]
76.05 vs 75.44
- AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models
AMQ beats BitStack · Avg. [Llama 2 70B, 3.5-bit]
77.11 vs 76.50
Newer alternatives
Recent methods in the same sub-problem, not yet superseded in the knowledge base.