Claude 63a6824881 PROF=1: per-request reports in mux serve mode too (the web-dashboard path) ...

run_serve_mux (SERVE_BATCH, used by openai_server.py / coli web) completes
requests in mux_done, so the run_serve per-turn hook never fired there.
Snapshot the window where hits0 is taken, record batched-forward latency
around step_decode_batch, report on stderr at DONE. With KV_SLOTS>1 the
window shares batched forwards across slots — same convention as the
existing STAT hit%.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TCBNCciBaHea41QmidLUMn

2026-07-16 08:56:12 -04:00

..

scripts

Translate user-facing runtime output to English, machine prefixes preserved, + CLI output test (#67, #85)

2026-07-12 13:38:40 +02:00

tests

Merge main into dev: Windows fixes (#275-279, #290) alongside engine work (#274, #293, #294, #297)

2026-07-16 07:59:53 +02:00

tools

bench: survive transient hub errors — retry with backoff, per-task isolation, atomic writes (#304)

2026-07-16 13:20:35 +03:00

.gitignore

Windows 11 native port, phase 1: MinGW-w64 static build, compat shims, setup + docs (#40)

2026-07-11 12:59:49 +02:00

backend_cuda.cu

GPU resident pipeline: batch CUDA attention, head-sharded kv_b, prefill expert groups, W4A16 mixed dispatch (#111)

2026-07-14 18:18:05 +02:00

backend_cuda.h

cuda(win): export + load the #111 pipeline ABI — 24 missing DLLEXPORT decorations + loader wrappers, unbreaks CUDA_DLL=1 (#201)

2026-07-14 18:43:56 +02:00

backend_loader.c

security(win): load coli_cuda.dll by absolute path, never from the CWD (DLL hijack)

2026-07-15 08:16:20 +02:00

backend_metal.h

Metal backend (Apple Silicon): batched experts + fused attention on GPU, unified-memory zero-copy, gated behind COLI_METAL — 2.06 tok/s M5 Max (#72, #87, #103)

2026-07-13 08:57:10 +02:00

backend_metal.mm

Metal backend (Apple Silicon): batched experts + fused attention on GPU, unified-memory zero-copy, gated behind COLI_METAL — 2.06 tok/s M5 Max (#72, #87, #103)

2026-07-13 08:57:10 +02:00

build_cuda.bat

windows-dev: CUDA build script, expert budget + grouped quant research

2026-07-15 02:32:12 -04:00

coli

Merge pull request #309 from bokiko/fix/bench-fetch-robust

2026-07-16 12:38:05 +02:00

compat.h

windows: PIPE default ON + compat_fadvise WILLNEED cache-warmer (pread path)

2026-07-15 06:14:15 -04:00

decode_batch.h

Unify continuous batching + heterogeneous runtime: decode batching, physical-core planning, disjoint VRAM/RAM placement, topp-policy warning (CPU-validated, CUDA on 6x5090) (#68)

2026-07-13 14:30:36 +02:00

doctor.py

config: fix UTF-8 decode crash + null config-value TypeError on Windows (doctor/resource_plan) (#184)

2026-07-14 14:17:42 +02:00

download_fp8.py

Windows-dev: grouped quantization + mixed precision + expert budget + download tool

2026-07-15 02:32:12 -04:00

glm.c

PROF=1: per-request reports in mux serve mode too (the web-dashboard path)

2026-07-16 08:56:12 -04:00

grammar.h

Translate user-facing runtime output to English, machine prefixes preserved, + CLI output test (#67, #85)

2026-07-12 13:38:40 +02:00

iobench.c

Windows: direct I/O (FILE_FLAG_NO_BUFFERING, 1.47x decode) + VirtualLock pin wiring + select_ctx device cache (#162)

2026-07-14 16:12:15 +02:00

json.h

security: harden safetensors/JSON parsers against malicious model files

2026-07-15 08:13:20 +02:00

Makefile

Makefile: add -std=gnu++17 to METALXX for Apple clang 16

2026-07-16 18:11:45 +08:00

olmoe.c

FreeBSD compatibility: platform guards in glm.c/olmoe.c (#206)

2026-07-14 22:08:12 +02:00

openai_server.py

server: clamp max_tokens to the server cap instead of rejecting it (fixes #260)

2026-07-15 14:56:01 +02:00

ref_glm.json

colibrì: pure-C GLM-5.2 (744B MoE) engine with disk-streamed experts

2026-07-05 20:57:25 +02:00

resource_plan.py

coli: measured Windows launcher defaults — OMP tuning parity + DIRECT/PIPE/PILOT_REAL

2026-07-15 12:06:48 -05:00

schema_gbnf.h

SCHEMA=<file.json>: JSON-Schema -> GBNF compiler for grammar-forced drafts (#148)

2026-07-14 16:31:09 +02:00

setup.sh

Translate user-facing runtime output to English, machine prefixes preserved, + CLI output test (#67, #85)

2026-07-12 13:38:40 +02:00

st.h

security: harden safetensors/JSON parsers against malicious model files

2026-07-15 08:13:20 +02:00

tier.h

Unify continuous batching + heterogeneous runtime: decode batching, physical-core planning, disjoint VRAM/RAM placement, topp-policy warning (CPU-validated, CUDA on 6x5090) (#68)

2026-07-13 14:30:36 +02:00

tok_unicode.h

Translate user-facing runtime output to English, machine prefixes preserved, + CLI output test (#67, #85)

2026-07-12 13:38:40 +02:00

tok.h

Translate user-facing runtime output to English, machine prefixes preserved, + CLI output test (#67, #85)

2026-07-12 13:38:40 +02:00

uring.h

use async and fix workers

2026-07-15 12:42:13 +02:00

warmup.ps1

Windows native port: serve-mode pipe fix + RAM detection + POSIX guards, AVX-VNNI kernel, gated CUDA DLL (#131, fixes #123)

2026-07-13 20:54:30 +02:00