code wiki / (root) / nx_nofloat_qwen_fastgen_gate.nx

nx_nofloat_qwen_fastgen_gate.nx

buildroot/runtime/nx_nofloat_qwen_fastgen_gate.nx

10910 B159 linesdepth 8pulls 19 transitivereach 0 importersview sourcekind gate/prooftopic nofloat
docsdependenciesstructsconstsfunctions

about

nx_nofloat_qwen_fastgen_gate.nx -- KV-cached generation with a DEQUANT-ONCE weight+head cache. The dqprobe gate MEASURED that ~88% of a decode step is re-dequantizing frozen weights (1213/1363 ms), and the streamed LM head re-dequantizes all 152K rows every token (~650 ms). Both are removed here by dequanting ONCE up front (nf_dequant_all_layers + nf_dequant_head_all) and reusing the cache every step. This is LOSSLESS: identical dequant, identical accumulation order -> the tokens MUST still be [12095,13,1084,374] (=' Paris' '.' ' It' ' is'), pinned as teeth. Reports one-time cache cost + per-token decode/head split. No hw writes (Rule 26). expect_exit: 0 license_tier: ORIGINAL

dependencies 10 imports · 0 importers

nx_syscalls.nx nx_tier.nx nx_le.nx nx_tensor.nx nx_gguf.nx nx_gguf_load.nx nx_gguf_meta.nx nx_nofloat_llm.nx nx_nofloat_tok.nx nx_gate_verdict.nx nx_nofloat_qwen_fastgen_gate.n

imports: nx_syscalls.nxnx_tier.nxnx_le.nxnx_tensor.nxnx_gguf.nxnx_gguf_load.nxnx_gguf_meta.nxnx_nofloat_llm.nxnx_nofloat_tok.nxnx_gate_verdict.nx

imported by: nobody (leaf or entry point)

call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown

main fg_puts sys_write sys_mmap sys_read_file sys_openat_rd sys_lseek sys_mmap ↻ sys_read sys_close nx_gguf_parse nx_le_read_u32 nx_le_read_u64 nx_le_read_u32 ↻ _gguf_skip_value nx_le_read_u64 ↻ nx_le_read_u32 ↻ _gguf_skip_value ↻ sys_mmap ↻ nx_gguf_meta_find nx_le_read_u64 ↻ nx_le_read_u32 ↻ _gmeta_bytes_equal nx_gguf_skip_value _gguf_skip_value ↻ fg_slen nx_gguf_meta_array_count nx_le_read_u64 ↻ nx_gguf_meta_array_first_e nx_gguf_find_tensor nx_gguf_tensor_at nx_gguf_name_equals nx_gguf_tensor_at ↻ load_named_q16 nx_gguf_find_tensor ↻ nx_gguf_tensor_at ↻ nx_gguf_tensor_n_values dequant_to_q16 f32_to_q16 nx_le_read_u32 ↻

structs

none

consts

none

functions

19func fg_puts(s: *u8) -> i64 { var n: i64=0; while s[n]!=(0 as u8){n=n+1} sys_write(1,s,n); return 0 }
called by 1: main calls 1: sys_write
20func fg_num(v: i64) -> i64 { let b: *u8=sys_mmap(28); var m: i64=v; if m<0{m=0-m;sys_write(1,"-" as *u8,1)} let t: *u8=sys_mmap(28); var k: i64=0; if m==0{t[0]=48 as u8;k=1} while m>0{t[k]=(48+(m%10)) as u8;m=m/10;k=k+1} var i: i64=0; while i<k{b[i]=t[k-1-i];i=i+1} sys_write(1,b,k); return 0 }
called by 1: main calls 2: sys_mmapsys_write
21func fg_slen(s: *u8) -> i64 { var n: i64=0; while s[n]!=(0 as u8){n=n+1} return n }
called by 1: main
23func main() -> i64