index
:
ffwd
master
Fastforward embedding inference engine
about
summary
refs
log
tree
commit
diff
log msg
author
committer
range
Commit message (
Collapse
)
Author
Age
Files
Lines
...
*
build: add gpu target for auto-selecting backend
Onur Gündüz
2026-06-14
2
-204
/
+35
|
*
chore: fix static analyzer warnings in tests/bench
Onur Gündüz
2026-06-14
3
-22
/
+47
|
*
server: add late-interaction rerank API
Onur Gündüz
2026-06-14
3
-48
/
+721
|
*
server: tune admission deadline by backend
Onur Gündüz
2026-06-14
1
-2
/
+5
|
*
server: make microbatch admission deterministic
Onur Gündüz
2026-06-14
2
-30
/
+131
|
*
build: add test-safetensors to PHONY targets
Onur Gündüz
2026-06-14
1
-2
/
+2
|
*
build: add test-bf16 and test-safetensors targets to Makefile
Onur Gündüz
2026-06-14
1
-1
/
+12
|
*
tools: add tools for model conversion/inspection and ranking embeddings
Onur Gündüz
2026-06-14
3
-0
/
+377
|
*
chore: rename project
Onur Gündüz
2026-06-14
22
-814
/
+805
|
*
tools: add benchmark comparison script
Onur Gündüz
2026-06-14
1
-0
/
+127
|
*
chore: comments housekeeping
Onur Gündüz
2026-06-14
5
-24
/
+17
|
*
bench: remove the legacy bench script
Onur Gündüz
2026-06-14
1
-500
/
+0
|
*
cuda: restore BF16 attention internals, length-route batched attention
Onur Gündüz
2026-06-14
1
-42
/
+130
|
*
dist: remove distributed layer-sharding (protocol, CLI flags, slice APIs)
Onur Gündüz
2026-06-14
9
-1403
/
+19
|
*
cuda: drop BF16 attention internals, keep fused rope + batched F32 attention
Onur Gündüz
2026-06-14
1
-125
/
+42
|
*
cuda: pointer-array batched attention for equal-length micro-batches
Onur Gündüz
2026-06-14
1
-8
/
+167
|
*
cuda: fuse QK norm+rope launches, BF16 attention output path
Onur Gündüz
2026-06-14
1
-45
/
+99
|
*
build: fix test_cli -lm on Linux, annotate CUDA API exports
Onur Gündüz
2026-06-14
2
-12
/
+23
|
*
build: target the local GPU with -arch=native in the nvcc rule
Onur Gündüz
2026-06-14
1
-1
/
+3
|
*
tests: tokenizer decode/NFC/contraction/loader-edge coverage
Onur Gündüz
2026-06-14
2
-13
/
+215
|
*
build: ignore profraw files
Onur Gündüz
2026-06-14
1
-0
/
+1
|
*
build: export only the public API from the shared library
Onur Gündüz
2026-06-14
7
-77
/
+145
|
*
kernels: delete dead audio/causal-LM leftovers and unused exports
Onur Gündüz
2026-06-14
9
-865
/
+4
|
*
tests: threaded kernel paths, bf16 widen, shard payload accounting
Onur Gündüz
2026-06-14
4
-8
/
+150
|
*
bench: progress output during sampling, ignore .gitignore in dirty stamp
Onur Gündüz
2026-06-14
2
-3
/
+11
|
*
bench: regression microbenchmarks with benchstat compare
Onur Gündüz
2026-06-14
5
-1
/
+589
|
*
build: backend targets produce static and shared libraries, drop make lib
Onur Gündüz
2026-06-14
1
-20
/
+10
|
*
tests: contextual batch-vs-singleton parity in test_workspace
Onur Gündüz
2026-06-14
1
-0
/
+73
|
*
tests: cover the late-interaction API and the CLI binary
Onur Gündüz
2026-06-14
6
-37
/
+722
|
*
build: relative paths and richer HTML in make coverage
Onur Gündüz
2026-06-14
1
-7
/
+19
|
*
tests: cross-check generic kernels against the SIMD dispatch choice
Onur Gündüz
2026-06-14
1
-0
/
+75
|
*
tests: fix goto-over-initialization cleanup in test_workspace, analyze tests ↵
Onur Gündüz
2026-06-14
1
-6
/
+8
|
|
|
|
in gate
*
tests: make test_workspace hermetic, add it to make test
Onur Gündüz
2026-06-14
3
-7
/
+35
|
*
server: resolve contextual separator from vocab, test contextual path
Onur Gündüz
2026-06-14
5
-11
/
+154
|
*
tests: drive the HTTP request path in-process in test_server
Onur Gündüz
2026-06-14
4
-103
/
+548
|
*
tokenizer: ASCII class table for pretokenizer codepoint checks
Onur Gündüz
2026-06-14
2
-9
/
+51
|
*
build: add make coverage (clang source-based, llvm-cov)
Onur Gündüz
2026-06-14
2
-16
/
+75
|
*
build: rename make targets blas->cpu, mlx->metal, shared->lib
Onur Gündüz
2026-06-14
3
-26
/
+26
|
*
server: default micro-batch cap to 32; add scheduler sweep driver
Onur Gündüz
2026-06-14
1
-3
/
+6
|
*
server: do not idle-reap clients with requests in flight
Onur Gündüz
2026-06-14
1
-1
/
+4
|
*
late: batch MaxSim as grouped GEMM plus row-max reduction
Onur Gündüz
2026-06-14
1
-2
/
+60
|
*
cpu: move BF16 widening scratch into the workspace
Onur Gündüz
2026-06-14
4
-152
/
+61
|
*
tests: hermetic BPE tokenizer test with fast/slow path equality
Onur Gündüz
2026-06-14
3
-0
/
+191
|
*
tokenizer: integer-pair BPE merging with linked-list symbols
Onur Gündüz
2026-06-14
2
-5
/
+241
|
*
tests: add server encoder unit tests via EMBED_SERVER_TEST include
Onur Gündüz
2026-06-14
3
-1
/
+97
|
*
cuda: follow model file dtype for weight storage by default
Onur Gündüz
2026-06-14
1
-1
/
+15
|
*
tests: make safetensors and bf16-model tests hermetic, join make test
Onur Gündüz
2026-06-14
5
-25
/
+273
|
*
gitignore: bench binary
Onur Gündüz
2026-06-14
1
-0
/
+1
|
*
tests,bench: move generated C checkers into standing files
Onur Gündüz
2026-06-14
5
-1
/
+448
|
*
cli: update help text
Onur Gündüz
2026-06-14
1
-2
/
+0
|
[prev]
[next]