enne2/qwentts.cpp
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

logs

Browse Source
This commit is contained in:
Pascal
2026-07-20 19:27:39 +02:00
parent 53c84fe1be
commit 877d8e1389
12 changed files with 441 additions and 967 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/clone-CUDA0-F32.log
+29 -75
View File
@@ -1,9 +1,9 @@
[Qwen] qwentts.cpp bcac463 (2026-07-05)
[Qwen] qwentts.cpp 53c84fe (2026-07-20)
ggml_cuda_init: found 1 CUDA devices (Total VRAM: 97247 MiB):
Device 0: NVIDIA RTX PRO 6000 Blackwell Workstation Edition, compute capability 12.0, VMM: yes, VRAM: 97247 MiB
load_backend: loaded CUDA backend from /mnt/workspace/git/qwentts.cpp/build/libggml-cuda.so
ggml_vulkan: Found 1 Vulkan devices:
ggml_vulkan: 0 = NVIDIA RTX PRO 6000 Blackwell Workstation Edition (NVIDIA) | uma: 0 | fp16: 1 | bf16: 0 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: NV_coopmat2
ggml_vulkan: 0 = NVIDIA RTX PRO 6000 Blackwell Workstation Edition (NVIDIA) | uma: 0 | fp16: 1 | bf16: 0 | fp4: 0 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: NV_coopmat2
load_backend: loaded Vulkan backend from /mnt/workspace/git/qwentts.cpp/build/libggml-vulkan.so
load_backend: loaded CPU backend from /mnt/workspace/git/qwentts.cpp/build/libggml-cpu-zen4.so
[Load] Talker backend: CUDA0 (CPU threads: 16)
@@ -23,9 +23,9 @@ load_backend: loaded CPU backend from /mnt/workspace/git/qwentts.cpp/build/libgg
[DAC] Loaded: 4 blocks (strides 8/5/4/3), 24 kHz mono out, weights 142.9 MB
[WeightCtx] Loaded 2 tensors, 3.0 MB into backend
[Pipeline] Ready: hop 1920 samples @ 24000 Hz mono, 16 codebooks @ 12.5 Hz
[KVCache] Allocated: 28 layers, 8 KV heads, head_dim 128, max_seq_len 4096 -> 896 MB
[KVCache] Allocated: 5 layers, 8 KV heads, head_dim 128, max_seq_len 16 -> 0 MB
[Pipeline] Loaded: arch=1b7 variant=base tokenizer=qwen3_tts_tokenizer_12hz codebooks=16 speaker_encoder=deferred speakers=0 fa=on clamp_fp16=off
[KVCache] Allocated: 28 layers, 8 KV heads, head_dim 128, max_seq_len 4096, 1 sets -> 896 MB
[KVCache] Allocated: 5 layers, 8 KV heads, head_dim 128, max_seq_len 16, 1 sets -> 0 MB
[Pipeline] Loaded: arch=1b7 variant=base tokenizer=qwen3_tts_tokenizer_12hz codebooks=16 speaker_encoder=deferred speakers=0 fa=on clamp_fp16=off max_batch=1
[BPE] Loaded from GGUF: 151676 vocab, 151291 merges, eos_id=151643
[BPE] Registered 5 arch special tokens (total specials=6)
[WAV] Read buffer: 380672 samples, 22050 Hz, 1 ch, 16 bit
@@ -33,7 +33,7 @@ load_backend: loaded CPU backend from /mnt/workspace/git/qwentts.cpp/build/libgg
[Audio-Resample] Done: 380672 -> 414337 samples
[WeightCtx] Loaded 76 tensors, 45.8 MB into backend
[SpeakerEncoder] Loaded: enc_dim=2048 sr=24000 mel_dim=128 hidden=512 mfa=1536 asp_attn=128 se=128 scale=8
[Pipeline] Speaker encoder lazy loaded in 4 ms
[Pipeline] Speaker encoder lazy loaded in 17 ms
[Debug] mel-spk: [1618, 128] first4: -6.330339 -6.170074 -6.011195 -6.243161
[Debug] mel-hann: [1024] first4: 0.000000 0.000009 0.000038 0.000085
[Debug] mel-basis: [128, 513] first4: 0.000000 0.033550 0.008569 0.000000
@@ -51,7 +51,7 @@ load_backend: loaded CPU backend from /mnt/workspace/git/qwentts.cpp/build/libgg
[EncDownsample] Loaded: k=4 stride=2, 512 -> 512 channels, weights 2.0 MB
[WeightCtx] Loaded 20 tensors, 34.0 MB into backend
[EncQuantizer] Loaded: 16 codebooks (1 semantic + 15 acoustic), 2048 entries x 256 dim, hidden 512, weights 34.0 MB
[Pipeline] Codec encoder lazy loaded in 31 ms
[Pipeline] Codec encoder lazy loaded in 90 ms
[Debug] audio-input: [412800] first4: -0.000546 -0.001162 -0.000871 0.000340
[Debug] seanet-out: [430, 512] first4: -0.134336 13.731120 -0.351051 3.013485
[Debug] enc-transformer-out: [430, 512] first4: 0.525800 3.116711 -0.187038 0.631811
@@ -76,78 +76,32 @@ load_backend: loaded CPU backend from /mnt/workspace/git/qwentts.cpp/build/libgg
[Debug] talker-hidden-prefill-l27: [225, 2048] first4: 7.131224 -34.471714 -7.108000 28.674042
[Debug] talker-hidden-prefill-final: [225, 2048] first4: 0.242833 -1.225771 -0.224906 0.890001
[Debug] talker-logits-prefill: [3072] first4: 1.171040 -7.796190 -6.822366 -10.286754
[Batch] Admit slot=0 T_ctx=225 prefill=45.0 ms build=47.6 ms (stall for 0 active slots)
[Sample] step=0 c0=27 u=-1.0000000000 subseq=0
[Sample-CP] g=0 c=1408 u=-1.0000000000 subseq=1
[Sample-CP] g=1 c=1824 u=-1.0000000000 subseq=2
[Sample-CP] g=2 c=400 u=-1.0000000000 subseq=3
[Sample-CP] g=3 c=240 u=-1.0000000000 subseq=4
[Sample-CP] g=4 c=2010 u=-1.0000000000 subseq=5
[Sample-CP] g=5 c=196 u=-1.0000000000 subseq=6
[Sample-CP] g=6 c=1984 u=-1.0000000000 subseq=7
[Sample-CP] g=7 c=816 u=-1.0000000000 subseq=8
[Sample-CP] g=8 c=684 u=-1.0000000000 subseq=9
[Sample-CP] g=9 c=1702 u=-1.0000000000 subseq=10
[Sample-CP] g=10 c=273 u=-1.0000000000 subseq=11
[Sample-CP] g=11 c=798 u=-1.0000000000 subseq=12
[Sample-CP] g=12 c=593 u=-1.0000000000 subseq=13
[Sample-CP] g=13 c=85 u=-1.0000000000 subseq=14
[Sample-CP] g=14 c=541 u=-1.0000000000 subseq=15
[Debug] codes-step0: [16] first4: 27.000000 1408.000000 1824.000000 400.000000
[Debug] next-emb-step0: [2048] first4: 0.003672 0.000908 0.039506 0.094594
[Debug] talker-hidden-step1: [2048] first4: 0.754625 2.407892 -0.056593 0.158147
[Sample] step=1 c0=1902 u=-1.0000000000 subseq=16
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=0 c=1924 u=-1.0000000000 subseq=17
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=1 c=2025 u=-1.0000000000 subseq=18
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=2 c=324 u=-1.0000000000 subseq=19
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=3 c=1501 u=-1.0000000000 subseq=20
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=4 c=1252 u=-1.0000000000 subseq=21
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=5 c=1831 u=-1.0000000000 subseq=22
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=6 c=1544 u=-1.0000000000 subseq=23
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=7 c=1058 u=-1.0000000000 subseq=24
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=8 c=1673 u=-1.0000000000 subseq=25
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=9 c=541 u=-1.0000000000 subseq=26
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=10 c=1162 u=-1.0000000000 subseq=27
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=11 c=729 u=-1.0000000000 subseq=28
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=12 c=781 u=-1.0000000000 subseq=29
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=13 c=85 u=-1.0000000000 subseq=30
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Sample-CP] g=14 c=1435 u=-1.0000000000 subseq=31
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Pipeline] Generated 8 frames
[Pipeline] Generated 16 frames
[Pipeline] Generated 24 frames
[Pipeline] Generated 32 frames
ggml_backend_cuda_graph_compute: CUDA graph warmup reset
ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Pipeline] Generated 40 frames
[Pipeline] Generated 48 frames
[Pipeline] Generated 56 frames
[Pipeline] Generated 64 frames
[Pipeline] Generated 8 frames (slot 0)
[Pipeline] Generated 16 frames (slot 0)
[Pipeline] Generated 24 frames (slot 0)
[Pipeline] Generated 32 frames (slot 0)
[Pipeline] Generated 40 frames (slot 0)
[Pipeline] Generated 48 frames (slot 0)
[Pipeline] Generated 56 frames (slot 0)
[Pipeline] Generated 64 frames (slot 0)
[Pipeline] Generation done : 64 frames
[Debug] codes-full: [64, 16] first4: 27.000000 1408.000000 1824.000000 400.000000
[Debug] output-audio: [122880] first4: 0.000739 0.000593 -0.000747 -0.000435
[Perf] PromptBuild 1.3 ms
[Perf] Prefill 27.4 ms (T_ctx prefill)
[Perf] TTFA 37.4 ms (first frame codes)
[Perf] TalkerDecode 304.7 ms (64 frames, 4.76 ms/frame)
[Perf] CodePredictor 389.8 ms (6.09 ms/frame)
[Perf] PromptBuild 47.6 ms
[Perf] Prefill 45.0 ms (T_ctx prefill)
[Perf] TTFA 1027.4 ms (first frame codes)
[Perf] TalkerDecode 608.3 ms (64 frames, 9.50 ms/frame)
[Perf] CodePredictor 1654.1 ms (25.84 ms/frame)
[Perf] HostCompose 0.1 ms (c0 sample + next emb)
[Perf] CodecDecode 14.8 ms
[Perf] Total 739.3 ms (64 frames, 10.85 ms/frame AR, audio 5.12 s, RTF 0.144)
[Perf] CodecDecode 24.8 ms
[Perf] Total 3261.0 ms (64 frames, 35.35 ms/frame AR, audio 5.12 s, RTF 0.637)
[WAV] Wrote cpp/clone/clone-cpp.wav: 122880 samples, 24000 Hz, mono S16
[Pipeline] Wrote 122880 samples (5.12 s) -> cpp/clone/clone-cpp.wav
[Input] Prompt: 100 chars: qwentts.cpp is a minimal C++17 and GGML port of Qwen3-TTS fo...
@@ -169,15 +123,15 @@ ggml_backend_cuda_graph_compute: CUDA graph warmup complete
[Cossim] PromptIDs exact: 100.00% (38 values)
[Cossim] RefCodes exact: 82.41% (3440 values)
[Cossim] Embed cos: 0.972420 max: 1.7531e-01 mean: 8.3100e-03
[Cossim] TrailingText cos: 1.000000 max: 2.9802e-08 mean: 5.1686e-10
[Cossim] TTSPadEmbed cos: 1.000000 max: 2.9802e-08 mean: 5.1686e-10
[Cossim] TrailingText cos: 1.000000 max: 2.9802e-08 mean: 6.8263e-10
[Cossim] TTSPadEmbed cos: 1.000000 max: 2.9802e-08 mean: 6.8263e-10
[Cossim] L0 cos: 0.986912 max: 6.6445e-01 mean: 1.1103e-02
[Cossim] L7 cos: 0.999992 max: 1.2570e+00 mean: 2.7510e-02
[Cossim] L14 cos: 0.999973 max: 2.1782e+00 mean: 5.9937e-02
[Cossim] L21 cos: 0.999719 max: 8.3606e+00 mean: 2.2268e-01
[Cossim] L27 cos: 0.998971 max: 2.3989e+01 mean: 5.1174e-01
[Cossim] Final cos: 0.994704 max: 7.3140e+00 mean: 1.6649e-01
[Cossim] Logits cos: 0.999743 max: 5.7110e-01 mean: 8.8929e-02
[Cossim] L21 cos: 0.999719 max: 8.3600e+00 mean: 2.2268e-01
[Cossim] L27 cos: 0.998971 max: 2.3987e+01 mean: 5.1173e-01
[Cossim] Final cos: 0.994704 max: 7.3135e+00 mean: 1.6649e-01
[Cossim] Logits cos: 0.999743 max: 5.7110e-01 mean: 8.8930e-02
[Cossim] NextEmbStep0 cos: 0.829216 max: 1.2523e-01 mean: 2.9129e-02
[Cossim] TalkerHiddenStep1 cos: 0.993854 max: 1.4329e+00 mean: 2.4202e-01
[Cossim] MelHann cos: 1.000000 max: 1.7881e-07 mean: 3.2043e-08