Vincenzo Fornaro 880cfb4ec5 Merge #192: serve stage 2 — response_format, per-request grammars, grammar-forced drafts ...

Per-request grammar constraints in serve mode: response_format plumbed through the API, per-request grammar compilation, and grammar-forced draft verification. Verified locally on top of current dev: clean build, token-exact tiny models unchanged, decode-batch helper tests ok, 40/40 openai_server tests. CI 8/8 green. Thanks @fabio-rovai for keeping this rebased through the refactor.

2026-07-20 21:09:04 +02:00

..

audit_win_shims.c

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

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

bench_dsa_select.c

refactor: split glm.c into colibri.c + 4 header modules

2026-07-19 21:12:04 +08:00

bench_idot.c

bench: microbench for AVX-VNNI idot accumulator A/B (not a gate)

2026-07-19 23:09:51 +02:00

bench_tensor_core.cu

cuda+engine: full fmt=4 (grouped int4 gs=64) support + diagnostic harness

2026-07-20 13:04:57 -04:00

bench_topp.c

refactor: split glm.c into colibri.c + 4 header modules

2026-07-19 21:12:04 +08:00

README_efficiency.md

tests: efficiency suite — tiny-model regression gates + full-model optimization dossier

2026-07-20 10:49:07 -04:00

test_backend_cuda.cu

cuda+engine: full fmt=4 (grouped int4 gs=64) support + diagnostic harness

2026-07-20 13:04:57 -04:00

test_backend_metal.mm

metal: parallel top-8 expert selection (r_top8_par), default ON

2026-07-19 17:26:56 -04:00

test_benchmark_cuda_fixture.py

profile effective CPU expert bandwidth

2026-07-18 05:34:41 +08:00

test_cli_output.py

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

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

test_compat_direct.c

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

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

test_decode_batch.c

serve stage 2: per-request grammars end-to-end — response_format -> SUBMIT -> grammar-forced drafts in the multiplexed server

2026-07-20 17:27:33 +01:00

test_doctor.py

doctor: platform-aware engine.binary check — Windows has no X_OK bit (#178, fixes #141)

2026-07-14 13:49:17 +02:00

test_dsa_select.c

refactor: split glm.c into colibri.c + 4 header modules

2026-07-19 21:12:04 +08:00

test_efficiency_report.py

tests: efficiency suite — tiny-model regression gates + full-model optimization dossier

2026-07-20 10:49:07 -04:00

test_env_defaults.py

win32: auto-enable the GPU in bare coli chat

2026-07-20 11:01:50 -04:00

test_grammar.c

Grammar-forced speculative drafts: GBNF grammar as a third draft source, guaranteed-accepted forced spans, lossless + opt-in (#48, #70)

2026-07-12 01:35:39 +02:00

test_grouped_g4_cuda.cu

cuda: grouped-int4 (fmt=4) support in the expert-group kernels (#334)

2026-07-20 08:56:49 +08:00

test_i4_acc512.c

expert dot: AVX-512 int4→float accumulator, runtime-switchable (I4_ACC512), +7% CPU-heavy decode, more accurate than AVX2 order (#95)

2026-07-13 15:24:25 +02:00

test_i4_grouped.c

refactor: split glm.c into colibri.c + 4 header modules

2026-07-19 21:12:04 +08:00

test_idot.c

refactor: split glm.c into colibri.c + 4 header modules

2026-07-19 21:12:04 +08:00

test_inefficiency.py

fix(tests): skip efficiency tests when glm_tiny fixture is absent (CI #360)

2026-07-20 10:49:07 -04:00

test_int3_convert.py

int3-g64 (fmt=5): per-group-scale 3-bit weight format — engine, converter, tests

2026-07-20 17:31:05 +01:00

test_int3_load.c

int3-g64 (fmt=5): per-group-scale 3-bit weight format — engine, converter, tests

2026-07-20 17:31:05 +01:00

test_int3.c

int3-g64 (fmt=5): per-group-scale 3-bit weight format — engine, converter, tests

2026-07-20 17:31:05 +01:00

test_iq3_pack.py

tests: skip the fmt=5 codec tests where numpy is absent

2026-07-21 01:34:26 +08:00

test_json.c

Local project checks and contribution templates: make check, dependency-free C/python tests, issue forms + blank issues, CONTRIBUTING (#20)

2026-07-10 08:56:41 +02:00

test_kv_alloc.c

refactor: split glm.c into colibri.c + 4 header modules

2026-07-19 21:12:04 +08:00

test_logit_nan.c

fix: update test_logit_nan to include colibri.c instead of glm.c

2026-07-19 21:18:33 +08:00

test_makefile_platform.py

refactor: split glm.c into colibri.c + 4 header modules

2026-07-19 21:12:04 +08:00

test_openai_server.py

review: docs section, 1 MiB grammar pre-check, negative tests, measured compile overhead

2026-07-20 17:27:33 +01:00

test_openai_tools_e2e.py

test: skip the tools e2e suite on Windows (shebang mock engine)

2026-07-19 11:01:26 +02:00

test_pipe_block.c

fix(build): test_pipe_block references glm.c, renamed to colibri.c by #391

2026-07-20 18:15:39 +02:00

test_pipe_cuda.cu

cuda+engine: full fmt=4 (grouped int4 gs=64) support + diagnostic harness

2026-07-20 13:04:57 -04:00

test_ragged_attention.cu

cuda: keep ragged KV resident across decode steps

2026-07-18 18:34:46 +08:00

test_resource_plan.py

fix(resource_plan): take last two lscpu fields, not [1]/[2] (#332 review)

2026-07-20 10:56:54 -04:00

test_router_cuda.cu

cuda: COLI_CUDA_ROUTER=1 — route the decode row on the layer's home device (#431 PR-A)

2026-07-20 02:53:43 +08:00

test_sample_nan.c

refactor: split glm.c into colibri.c + 4 header modules

2026-07-19 21:12:04 +08:00

test_schema_gbnf.c

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

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

test_st_pread.c

test_st_pread: Windows-compatible — relative tmpdir, fork subtest gated POSIX

2026-07-16 16:32:43 -04:00

test_st.c

Local project checks and contribution templates: make check, dependency-free C/python tests, issue forms + blank issues, CONTRIBUTING (#20)

2026-07-10 08:56:41 +02:00

test_stops.c

refactor: split glm.c into colibri.c + 4 header modules

2026-07-19 21:12:04 +08:00

test_tier.c

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

test_tok_o200k.c

test_tok_o200k: fgets instead of getline for the windows job

2026-07-16 16:33:22 -04:00

test_tok.c

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

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

test_topp.c

refactor: split glm.c into colibri.c + 4 header modules

2026-07-19 21:12:04 +08:00

test_uring.c

refactor: split glm.c into colibri.c + 4 header modules

2026-07-19 21:12:04 +08:00

tok_o200k_cases.txt

tests: o200k tokenizer coverage, no model download

2026-07-16 15:17:21 -04:00

tok_o200k_tiny.json

tests: o200k tokenizer coverage, no model download

2026-07-16 15:17:21 -04:00