code wiki / _hdl_build / nx_vram_frontier.nx

nx_vram_frontier.nx

buildroot/runtime/_hdl_build/nx_vram_frontier.nx

1492 B28 linesdepth 3pulls 3 transitivereach 1 importersview sourcekind librarytopic vram
docsdependenciesstructsconstsfunctions

about

nx_vram_frontier.nx -- MAP THE FRONTIER: the achievable set on a fixed consumer GPU is not set by the hardware, it is PUSHED OUTWARD by smarter math. For a budget, this walks a ladder of stacked techniques (fp16 -> Q4 weights -> +Q8 KV -> +GQA -> +Q4 K-cache) and shows, at each rung, (a) the LARGEST model that fits at a fixed context and (b) the LONGEST context that fits for a fixed model. Each rung should dominate the one below: same silicon, more capability. The deep-research adds further rungs (2-bit/BitNet weights, MLA, KV eviction, weight streaming). license_tier: ORIGINAL

dependencies 1 imports · 1 importers

nx_vram_budget.nx nx_vram_frontier.nx nx_vram_frontier_test.nx

imports: nx_vram_budget.nx

imported by: nx_vram_frontier_test.nx

structs

none

consts

none

functions

12func vfr_usable_mb(budget_gb: i64, act_mb: i64, overhead: i64) -> i64
called by 1: main
17func vfr_max_model_m(usable_mb: i64, kv_mb: i64, w_bits: i64) -> i64
called by 1: main
24func vfr_max_ctx(usable_mb: i64, weights_mb: i64, kv_at_ref_mb: i64, ref_ctx: i64) -> i64
called by 1: main