nx_f32_llm_live_gen_test.nx
buildroot/runtime/nx_f32_llm_live_gen_test.nx
about
nx_f32_llm_live_gen_test.nx -- LIVE-FIRE TEXT GENERATION from real
Qwen2.5-0.5B. Loads model + vocab + cache + runs greedy 5-token
generation from a fixed prompt. Writes output bytes to
/tmp/nx_qwen_gen.txt for inspection.
This is the substrate's "text in, text out" proof on real
production weights.
dependencies 21 imports · 0 importers
diagram shows first 10 each side; +11 more imports, +0 more importers in the complete lists below.
imports: nx_syscalls.nxnx_tier.nxnx_bpe.nxnx_gguf.nxnx_gguf_load.nxnx_gguf_meta.nxnx_f32.nxnx_f32_kv_cache.nxnx_f32_lazy_weight.nxnx_f32_llama_block.nxnx_f32_llama_block_v4.nxnx_f32_llama_stack_v4.nxnx_f32_llama_layer_lazy_load.nxnx_f32_llm.nxnx_f32_llm_v4.nxnx_f32_llm_read_dims.nxnx_f32_bpe_load.nxnx_f32_llm_special_tokens.nxnx_f32_sampler.nxnx_f32_llm_run_v2.nxnx_f32_llm_run_v3.nx
imported by: nobody (leaf or entry point)
call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown
structs
| none |
consts
| none |
functions
| 31 | func main() -> i64 |