feat(audio8_tts): Falcon-H1 0.1B (Mamba2 + attention) support and codec/AR performance - #444

Open

gqf2008 wants to merge 17 commits into

0xShug0:mainfrom

gqf2008:feat/audio8-tts-falcon-h1-01b

Commits on Sep 4, 2026

feat(audio8_tts): Falcon-H1 0.1B (Mamba2 + attention) stateful forward port
gqf2008
committed
993a9a2
fix: Falcon-H1 state corruption and numerical stability
gqf2008
committed
3c95e68
docs: record Falcon-H1 0.1B port status and open issues
gqf2008
committed
4455a7b
docs(audio8_tts): mark Falcon-H1 open issues resolved with root causes
gqf2008
committed
f5e71b0
perf(audio8_tts): run fast AR graph on a dedicated CPU backend
gqf2008
committed
0ca4a58
feat(audio8_tts): time codec decode graph build vs compute
gqf2008
committed
9498828
perf(conv): Metal fast path for stride-1 conv1d on time-fast layouts
gqf2008
committed
846a056
perf(audio8_tts): run Falcon-H1 per-token step graph on the CPU backend
gqf2008
committed
ff7ebf3
perf(audio8_tts): keep Falcon KV cache in padded host buffers, write slots in-graph
gqf2008
committed
715b1f9
perf(audio8_tts): reuse the Falcon step graph per KV capacity bucket
gqf2008
committed
1e961ff
perf(audio8_tts): accumulate conv1d per-tap GEMMs in-place via ggml_mul_mat_acc
gqf2008
committed
9184efc
perf: fused snake1d activation op (GGML_OP_SNAKE_1D) for audio8_tts codec
gqf2008
committed
94bf6a1
perf(audio8_tts): zero-copy Falcon step state and constants on CPU backend
gqf2008
committed
dbe68ad
perf(audio8_tts): run decoder blocks in channel-fast layout on Metal
gqf2008
committed
689aab1
fix(ggml-metal): ssm_scan reduction reads garbage when sgptg < NW
gqf2008
committed
771b617
fix(audio8_tts): Falcon-H1 0.1B logits parity and long-run stability
gqf2008
committed
e2bd82f

Commits on Sep 15, 2026

Merge commit 'fcba3a4' into feat/audio8-tts-falcon-h1-01b
gqf2008
committed
b6c8458

Navigation Menu

feat(audio8_tts): Falcon-H1 0.1B (Mamba2 + attention) support and codec/AR performance - #444

feat(audio8_tts): Falcon-H1 0.1B (Mamba2 + attention) support and codec/AR performance#444
gqf2008 wants to merge 17 commits into
0xShug0:mainfrom
gqf2008:feat/audio8-tts-falcon-h1-01b

Commits on Sep 4, 2026

feat(audio8_tts): Falcon-H1 0.1B (Mamba2 + attention) stateful forward port

fix: Falcon-H1 state corruption and numerical stability

docs: record Falcon-H1 0.1B port status and open issues

docs(audio8_tts): mark Falcon-H1 open issues resolved with root causes

perf(audio8_tts): run fast AR graph on a dedicated CPU backend

feat(audio8_tts): time codec decode graph build vs compute

perf(conv): Metal fast path for stride-1 conv1d on time-fast layouts

perf(audio8_tts): run Falcon-H1 per-token step graph on the CPU backend

perf(audio8_tts): keep Falcon KV cache in padded host buffers, write slots in-graph

perf(audio8_tts): reuse the Falcon step graph per KV capacity bucket

perf(audio8_tts): accumulate conv1d per-tap GEMMs in-place via ggml_mul_mat_acc

perf: fused snake1d activation op (GGML_OP_SNAKE_1D) for audio8_tts codec

perf(audio8_tts): zero-copy Falcon step state and constants on CPU backend

perf(audio8_tts): run decoder blocks in channel-fast layout on Metal

fix(ggml-metal): ssm_scan reduction reads garbage when sgptg < NW

fix(audio8_tts): Falcon-H1 0.1B logits parity and long-run stability

Commits on Sep 15, 2026

Merge commit 'fcba3a4' into feat/audio8-tts-falcon-h1-01b