{ "variant": "fp32", "recheck": false, "model_load_seconds": 0.7284324999927776, "conditioning_seconds": 0.13869441598944832, "reference_sha256": "ace8518ab1d3b45e702a4c54700cab378a3d6e8172dfa8cd3dd51d8e176cfac3", "parameters": { "exaggeration": 0.5, "cfg_weight": 0.5, "temperature": 0.8, "repetition_penalty": 1.2, "min_p": 0.05, "top_p": 0.95, "max_new_tokens": 400, "verbose": false }, "runtime_quantized_matrices": [], "versions": { "mlx": "0.32.3", "mlx-audio": "0.5.7", "numpy": "2.4.6", "soundfile": "0.14.0", "huggingface-hub": "1.33.0" }, "warm_repetitions": [ { "file": "fp32-warm-0.wav", "sha256": "37c7509a84deb7da0036be7e61d74e06c10c1bd5025ed56c94b2b1a6b1763995", "float_sha256": "7afadd7bb8cd77e3bfec39a942a0a1069eb628ce432e883935c1dee1a77f1bac", "generation_seconds": 7.8654603750037495, "audio_seconds": 8.48, "rtf": 0.9275307045994987, "mlx_peak_bytes": 4752006708, "rms": 0.06957667850218721, "peak": 0.6255986094474792, "clipped_fraction": 0.0, "repeat": 0, "first_generation": true, "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "lang": "en", "seed": 20261009 }, { "file": "fp32-warm-1.wav", "sha256": "d1deef4c4bdbed660f4766cdec6bed39df85223b6f2f1ba16bf3ed9f1502ced3", "float_sha256": "826671d540a9158954fbbf9b5920c4e680888b136a85ae6aa08e53a7831d5c80", "generation_seconds": 7.854540125001222, "audio_seconds": 8.48, "rtf": 0.926242939269012, "mlx_peak_bytes": 4752006708, "rms": 0.06957667849927777, "peak": 0.6255986094474792, "clipped_fraction": 0.0, "repeat": 1, "first_generation": false, "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "lang": "en", "seed": 20261009 }, { "file": "fp32-warm-2.wav", "sha256": "5a6eee18610c68394f9595eee40a2a9eedb0ad2fdd75a668a35dce450a87a4e5", "float_sha256": "dccc883aaf8a6cc24085819ff7d5bc1452199c247c7643ceaae1ec947eeffa64", "generation_seconds": 7.980229666005471, "audio_seconds": 8.48, "rtf": 0.9410648191044186, "mlx_peak_bytes": 4752006708, "rms": 0.06957667851759584, "peak": 0.6255986094474792, "clipped_fraction": 0.0, "repeat": 2, "first_generation": false, "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "lang": "en", "seed": 20261009 }, { "file": "fp32-warm-3.wav", "sha256": "03503a2eea73077f40ffdf32d32df66dafc871584bd5248170fa34eb5cd22b03", "float_sha256": "beb525e06f41ad919df47586df37343559dbb2ecbf61eb423cc8ab0aadc5a9b4", "generation_seconds": 7.764641332993051, "audio_seconds": 8.48, "rtf": 0.915641666626539, "mlx_peak_bytes": 4752006708, "rms": 0.06957667854260867, "peak": 0.6255986094474792, "clipped_fraction": 0.0, "repeat": 3, "first_generation": false, "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "lang": "en", "seed": 20261009 }, { "file": "fp32-warm-4.wav", "sha256": "49e24be404cf26834b36cc7c400fa0ca3ccb1e93d37af30fd41ec22c864c87e1", "float_sha256": "ac87484e8c766519510dc4d43616b71e3d8fdb1b2aaf7406636371c86c2580ae", "generation_seconds": 7.7425289589882595, "audio_seconds": 8.48, "rtf": 0.913034075352389, "mlx_peak_bytes": 4752006708, "rms": 0.06957667853388305, "peak": 0.6255986094474792, "clipped_fraction": 0.0, "repeat": 4, "first_generation": false, "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "lang": "en", "seed": 20261009 }, { "file": "fp32-warm-5.wav", "sha256": "e69091505a0ef7f8c578804cc1a70d1063b5581fa57713034d1a5a1481d0f05f", "float_sha256": "3b65121db5204667624630221a5cabff74e3d9d92dcfd42720d2cd8aedc104f9", "generation_seconds": 8.44908179200138, "audio_seconds": 8.48, "rtf": 0.996353984905823, "mlx_peak_bytes": 4752006708, "rms": 0.06957667852567641, "peak": 0.6255986094474792, "clipped_fraction": 0.0, "repeat": 5, "first_generation": false, "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "lang": "en", "seed": 20261009 } ], "end_to_end": [ { "file": "fp32-e2e-0-20261009.wav", "sha256": "1daa7a18c4120688e2ead89aeb9589fb9b1723ade090edef30f112e3b18aecef", "float_sha256": "3b24613a454d05f95b4e55383aa7dc0745dc42b73b8345df4803721747027c4b", "generation_seconds": 8.922113583001192, "audio_seconds": 8.48, "rtf": 1.0521360357312726, "mlx_peak_bytes": 4752172604, "rms": 0.06957667853726353, "peak": 0.6255986094474792, "clipped_fraction": 0.0, "prompt": 0, "lang": "en", "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "seed": 20261009, "speech_tokens": 212, "speech_token_sha256": "1225f7a4fdc94204bc692e58fa0ea124684670fe74211411d007ad53eb4e4477", "timing_includes_token_capture": true }, { "file": "fp32-e2e-0-42.wav", "sha256": "5dc73656bc9ba5ffc4cab9d2fc77994802b74ae59cccad278f1652e6a5cfa171", "float_sha256": "d5bcf87ae59278579a926d6aef730c0976084682c1d2ce80994449f0be16d9f5", "generation_seconds": 8.271469291998073, "audio_seconds": 7.84, "rtf": 1.055034348469142, "mlx_peak_bytes": 4692263460, "rms": 0.06993587318044066, "peak": 0.5513496398925781, "clipped_fraction": 0.0, "prompt": 0, "lang": "en", "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "seed": 42, "speech_tokens": 196, "speech_token_sha256": "c1c2de8f00a7122c5becec088297c2236ea36df1c2c2c744447d65123736f0ff", "timing_includes_token_capture": true }, { "file": "fp32-e2e-1-20261009.wav", "sha256": "cd99a4335924667403cbc604917465e4c8816bfb04a319788c414c3c89fc763f", "float_sha256": "94c68ff8713b7ab84c97bf0093ec4e1a8ca1010b72e619c4080a81b9f8298dfd", "generation_seconds": 6.114699500001734, "audio_seconds": 5.8, "rtf": 1.0542585344830577, "mlx_peak_bytes": 4646591052, "rms": 0.0800175289064696, "peak": 0.631700873374939, "clipped_fraction": 0.0, "prompt": 1, "lang": "en", "text": "Hello mate, the training checkpoint is saved, and this is Chatterbox speaking locally on the Mac.", "seed": 20261009, "speech_tokens": 145, "speech_token_sha256": "5cfc24860df526a94926fb5c5e241ad4ebf77207f4f4f9c1008c90f6f7a15df9", "timing_includes_token_capture": true }, { "file": "fp32-e2e-1-42.wav", "sha256": "f338c538651c3d754a206a26d62cf75e5cc9dac3ad2cb45df9d07ad3f4c3082b", "float_sha256": "3ddee0e6e7336353bb96677f7ee4ec205b6e909f4298bf07436a90ce6ab20f81", "generation_seconds": 5.884816999998293, "audio_seconds": 5.6, "rtf": 1.0508601785711238, "mlx_peak_bytes": 4619539828, "rms": 0.08179977043562504, "peak": 0.562351405620575, "clipped_fraction": 0.0, "prompt": 1, "lang": "en", "text": "Hello mate, the training checkpoint is saved, and this is Chatterbox speaking locally on the Mac.", "seed": 42, "speech_tokens": 140, "speech_token_sha256": "fef7f0074299cbb05c480ebb87e82d76eb57335e55e6820a40d0d524b6675496", "timing_includes_token_capture": true }, { "file": "fp32-e2e-2-20261009.wav", "sha256": "f0e02f34a3345d3dbd3c6782df1f4e98f5778ba38e213dfcacddaaf9ef3d27b1", "float_sha256": "cb3c8ede8dd9347f7e865edd0d0ae638cfb2379b45dd7fe4c0d376cc168d3303", "generation_seconds": 5.93163387500681, "audio_seconds": 5.72, "rtf": 1.0369989291969948, "mlx_peak_bytes": 4636973148, "rms": 0.07540991201849923, "peak": 0.5489928126335144, "clipped_fraction": 0.0, "prompt": 2, "lang": "en", "text": "Please check the appointment on Thursday at three thirty. Bring the blue folder, your keys, and a bottle of water.", "seed": 20261009, "speech_tokens": 143, "speech_token_sha256": "0162792dd3cccdf3b23f649b2f317e777d5431be8b43c4b898534c3915ba5c30", "timing_includes_token_capture": true }, { "file": "fp32-e2e-2-42.wav", "sha256": "e53591d240245ecc58deb1ba88e57c7421ce06dbc58617797b8e0a1c94192dfd", "float_sha256": "05a40fa6566ab897c76a29fd8d3cc6e41be72ab3ee1a430d0aebefa39a31c9b6", "generation_seconds": 5.530443750001723, "audio_seconds": 5.32, "rtf": 1.0395570958649856, "mlx_peak_bytes": 4582723756, "rms": 0.07560245657977253, "peak": 0.5662506222724915, "clipped_fraction": 0.0, "prompt": 2, "lang": "en", "text": "Please check the appointment on Thursday at three thirty. Bring the blue folder, your keys, and a bottle of water.", "seed": 42, "speech_tokens": 133, "speech_token_sha256": "13365eea735200dba4ce5afe9e981a1dd747ae44e67036cbe8704b23cc68834c", "timing_includes_token_capture": true }, { "file": "fp32-e2e-3-20261009.wav", "sha256": "42d1aa551f2f5d6faec832db47a383865722baea6548948c474a31638c07e53e", "float_sha256": "f4c7868bdc888a7c4e1735780673013afce7b14c1bd13e2b67daf9ac35809f40", "generation_seconds": 6.33472037500178, "audio_seconds": 6.52, "rtf": 0.9715828796015, "mlx_peak_bytes": 4748307388, "rms": 0.09043996701881027, "peak": 0.663648784160614, "clipped_fraction": 0.0, "prompt": 3, "lang": "es", "text": "Este es el checkpoint de Chatterbox Turbo de oMLX. Fue cuantizado a partir del modelo original fp16.", "seed": 20261009, "speech_tokens": 163, "speech_token_sha256": "912cde02e0ae66e34fefdca7455d29fe0a3a88094574e2fea2f3af2d35252004", "timing_includes_token_capture": true }, { "file": "fp32-e2e-3-42.wav", "sha256": "a1d33c86962950a687051472cd5986b12f79dfccaf4aeac3952f8cd1e933512a", "float_sha256": "70c8a011b449abe2fa4a73f1190bd564922e53a66eed70ca83ba9f03a55b3fa8", "generation_seconds": 6.24799920800433, "audio_seconds": 6.4, "rtf": 0.9762498762506766, "mlx_peak_bytes": 4728548052, "rms": 0.08401720271030556, "peak": 0.6792764067649841, "clipped_fraction": 0.0, "prompt": 3, "lang": "es", "text": "Este es el checkpoint de Chatterbox Turbo de oMLX. Fue cuantizado a partir del modelo original fp16.", "seed": 42, "speech_tokens": 160, "speech_token_sha256": "510a36cc71321f20b5a06826b4884130d4adc8618585757ffdc2e74ec5899b9e", "timing_includes_token_capture": true }, { "file": "fp32-e2e-4-20261009.wav", "sha256": "c7627be7f910d3832180b846d59449c697e7ac24dcc91aa02f86d788ce0c25fc", "float_sha256": "00105c5c0e778d0559249582df233929df55968d267c077526be4c95b57df922", "generation_seconds": 7.8277415839984315, "audio_seconds": 6.08, "rtf": 1.287457497368163, "mlx_peak_bytes": 4684603668, "rms": 0.08884582979242805, "peak": 0.8400182127952576, "clipped_fraction": 0.0, "prompt": 4, "lang": "es", "text": "Hola, amigo, el checkpoint de entrenamiento está guardado y este es Chatterbox hablando localmente en la Mac.", "seed": 20261009, "speech_tokens": 152, "speech_token_sha256": "bbee0e53132295f6c223b590727068ebf1b10ddc1a98d20c9889f806c232e742", "timing_includes_token_capture": true }, { "file": "fp32-e2e-4-42.wav", "sha256": "7aee51659cfe07642c0aaf462ebd4276e05aec83dd135c03ded6bac11850d2b0", "float_sha256": "cd2a44f1d15453fe4a284a967f41a76b989b3670f3b929a61e7ff4138ab2e798", "generation_seconds": 7.983857915998669, "audio_seconds": 6.16, "rtf": 1.2960808305192644, "mlx_peak_bytes": 4694696708, "rms": 0.08318754580783018, "peak": 0.7967129945755005, "clipped_fraction": 0.0, "prompt": 4, "lang": "es", "text": "Hola, amigo, el checkpoint de entrenamiento está guardado y este es Chatterbox hablando localmente en la Mac.", "seed": 42, "speech_tokens": 154, "speech_token_sha256": "a1be4dc194c481b4c1dd62337f92e6bd9769fa6465784acc4cfa4f018131704e", "timing_includes_token_capture": true }, { "file": "fp32-e2e-5-20261009.wav", "sha256": "27e6289dbe6c807416bcd175fd5b1409e361dfc15337e12d2470a8f911403197", "float_sha256": "6569afc8177641d42a6fe46ba46990d7c6f491fa8b61c99fa6bff90d9eeb2761", "generation_seconds": 8.032594167001662, "audio_seconds": 6.08, "rtf": 1.321150356414747, "mlx_peak_bytes": 4684603668, "rms": 0.09924388267937967, "peak": 0.7688873410224915, "clipped_fraction": 0.0, "prompt": 5, "lang": "es", "text": "Por favor, revisa la cita del jueves a las tres y media. Trae la carpeta azul, tus llaves y una botella de agua.", "seed": 20261009, "speech_tokens": 152, "speech_token_sha256": "aa5fff77bc601f519ec9ba885eb0a4918aad4ef242fd076670f4ffef9fd50847", "timing_includes_token_capture": true }, { "file": "fp32-e2e-5-42.wav", "sha256": "5ed6ff05751d5c9f5cca8081e3ad04de2f371514dd9470f270e2ae5a36d70ed0", "float_sha256": "40ffa27fc34308e8ec1b874e86e7eb3b0e2c50b3c0c601156ac52aa7fb2f679c", "generation_seconds": 10.38800774999254, "audio_seconds": 8.28, "rtf": 1.2545903079701135, "mlx_peak_bytes": 4727718764, "rms": 0.08350096010153857, "peak": 0.9687581062316895, "clipped_fraction": 0.0, "prompt": 5, "lang": "es", "text": "Por favor, revisa la cita del jueves a las tres y media. Trae la carpeta azul, tus llaves y una botella de agua.", "seed": 42, "speech_tokens": 207, "speech_token_sha256": "0012d919bf529baf19b526f9ee344ad618f3963beb3bb616efac8b261443cd59", "timing_includes_token_capture": true } ], "controlled_decoder": [ { "file": "fp32-decoder-0-20261009.wav", "sha256": "2896406ba2c0455d740c2795ff10007ae500f2d8fdfb466a75a3350f13b3cec4", "float_sha256": "75958e6796f5ede5165b9e1a14ee45b54775600af0645ccb7cdaf19a5cd394aa", "generation_seconds": 4.182588665993535, "audio_seconds": 8.48, "rtf": 0.49322979551810553, "mlx_peak_bytes": 4752172788, "rms": 0.06957545279596994, "peak": 0.6253780722618103, "clipped_fraction": 0.0, "prompt": 0, "lang": "en", "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "seed": 20261009, "decoder_seed": 9182, "speech_tokens": 212, "source_token_sha256": "1225f7a4fdc94204bc692e58fa0ea124684670fe74211411d007ad53eb4e4477", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 8.940696716308594e-08, "self_repeat_relative_rms_delta": 4.7802608094116295e-08, "self_repeat_control": { "file": "fp32-decoder-self-0-20261009.wav", "sha256": "abb35602238cbd2050d3bf8df3d48464039cca4af696735137020ab89dea3461", "float_sha256": "c1fe4529c70ad7d0a0cd9e57bbcb90fa19884d7a1ed74308ab74e8c55931a9c9", "generation_seconds": 0.0, "audio_seconds": 8.48, "rtf": 0.0, "mlx_peak_bytes": 4752991988, "rms": 0.06957545274505651, "peak": 0.6253780722618103, "clipped_fraction": 0.0, "prompt": 0, "lang": "en", "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "seed": 20261009 } }, { "file": "fp32-decoder-0-42.wav", "sha256": "7951e081c833a66ee645e9ff5186d8c7fa303103dd3cb0a3de72665d226bdf68", "float_sha256": "d8f775abf14c848549bbc058376b57903672bee21f07e891992c23e50cc141eb", "generation_seconds": 3.886552790994756, "audio_seconds": 7.84, "rtf": 0.495733774361576, "mlx_peak_bytes": 4693070684, "rms": 0.06993799849085634, "peak": 0.5512809157371521, "clipped_fraction": 0.0, "prompt": 0, "lang": "en", "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "seed": 42, "decoder_seed": 9182, "speech_tokens": 196, "source_token_sha256": "c1c2de8f00a7122c5becec088297c2236ea36df1c2c2c744447d65123736f0ff", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 8.940696716308594e-08, "self_repeat_relative_rms_delta": 4.7897767530002966e-08, "self_repeat_control": { "file": "fp32-decoder-self-0-42.wav", "sha256": "0fe08075bfb16e8ac5e3fa165432d7ea0158738fe01d7dacf8522f922d1b46ce", "float_sha256": "70dbbed8a401793acc0764aadb9c8f4b0aad6c8eb34805a9addffd455544ef8e", "generation_seconds": 0.0, "audio_seconds": 7.84, "rtf": 0.0, "mlx_peak_bytes": 4693070684, "rms": 0.0699379985302829, "peak": 0.5512809157371521, "clipped_fraction": 0.0, "prompt": 0, "lang": "en", "text": "This is the oMLX Chatterbox Turbo checkpoint. It was quantized from the original fp16 model.", "seed": 42 } }, { "file": "fp32-decoder-1-20261009.wav", "sha256": "0fb373904f1e3800d309a4d07b16caef0521b7d70ff10e620b405b21777c6da0", "float_sha256": "c65224cbce9643b87852803ecfd7cdd5405ace02402c083ec5617b21c49eef18", "generation_seconds": 2.982861625001533, "audio_seconds": 5.8, "rtf": 0.5142864870692299, "mlx_peak_bytes": 4647331384, "rms": 0.08001458855372089, "peak": 0.6316264271736145, "clipped_fraction": 0.0, "prompt": 1, "lang": "en", "text": "Hello mate, the training checkpoint is saved, and this is Chatterbox speaking locally on the Mac.", "seed": 20261009, "decoder_seed": 9182, "speech_tokens": 145, "source_token_sha256": "5cfc24860df526a94926fb5c5e241ad4ebf77207f4f4f9c1008c90f6f7a15df9", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 8.940696716308594e-08, "self_repeat_relative_rms_delta": 4.746361170759883e-08, "self_repeat_control": { "file": "fp32-decoder-self-1-20261009.wav", "sha256": "50b58b7e0fc9919a78ca153a72b1d6dab09dcc475f264c7982445becd143a1f5", "float_sha256": "dc3c55f5c95f380f8610020cc5c3037b2dfe8b97dfacc304ad54478a82669094", "generation_seconds": 0.0, "audio_seconds": 5.8, "rtf": 0.0, "mlx_peak_bytes": 4647331384, "rms": 0.08001458854113612, "peak": 0.6316264271736145, "clipped_fraction": 0.0, "prompt": 1, "lang": "en", "text": "Hello mate, the training checkpoint is saved, and this is Chatterbox speaking locally on the Mac.", "seed": 20261009 } }, { "file": "fp32-decoder-1-42.wav", "sha256": "ff6919f4a45bc9baaa1fb5ceb72b2dd17ba22dd3b510a15ad3dfe94f5dc42638", "float_sha256": "404d43eabf4ab2c41c3cfabccba7fb10c157aa9b781a6b59c24572a0c34bcac7", "generation_seconds": 2.5094010409957264, "audio_seconds": 5.6, "rtf": 0.44810732874923687, "mlx_peak_bytes": 4620084224, "rms": 0.08180370749520352, "peak": 0.5624030828475952, "clipped_fraction": 0.0, "prompt": 1, "lang": "en", "text": "Hello mate, the training checkpoint is saved, and this is Chatterbox speaking locally on the Mac.", "seed": 42, "decoder_seed": 9182, "speech_tokens": 140, "source_token_sha256": "fef7f0074299cbb05c480ebb87e82d76eb57335e55e6820a40d0d524b6675496", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 8.940696716308594e-08, "self_repeat_relative_rms_delta": 5.10611428694574e-08, "self_repeat_control": { "file": "fp32-decoder-self-1-42.wav", "sha256": "3f72f37840b5c48ab5c9c67d6587af7cb9fda348f7713a11c554c1e4ad1f1981", "float_sha256": "f37648f1a8b58fe7b3e5069000a1d851e6947a4962eddcd3ff214aa953947690", "generation_seconds": 0.0, "audio_seconds": 5.6, "rtf": 0.0, "mlx_peak_bytes": 4620084224, "rms": 0.08180370748465525, "peak": 0.5624030828475952, "clipped_fraction": 0.0, "prompt": 1, "lang": "en", "text": "Hello mate, the training checkpoint is saved, and this is Chatterbox speaking locally on the Mac.", "seed": 42 } }, { "file": "fp32-decoder-2-20261009.wav", "sha256": "1869c7b4212901cb2f460c7f93d9317322769f38910933d3e78e68c5a83acd6e", "float_sha256": "516ab3df55c5daf4c7f73b2f9133413c857dbf275a9149ced02357a53c7f7fb7", "generation_seconds": 2.56840312499844, "audio_seconds": 5.72, "rtf": 0.44902152534937767, "mlx_peak_bytes": 4637501364, "rms": 0.07540756299804834, "peak": 0.549577534198761, "clipped_fraction": 0.0, "prompt": 2, "lang": "en", "text": "Please check the appointment on Thursday at three thirty. Bring the blue folder, your keys, and a bottle of water.", "seed": 20261009, "decoder_seed": 9182, "speech_tokens": 143, "source_token_sha256": "0162792dd3cccdf3b23f649b2f317e777d5431be8b43c4b898534c3915ba5c30", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 8.940696716308594e-08, "self_repeat_relative_rms_delta": 4.7621249166240887e-08, "self_repeat_control": { "file": "fp32-decoder-self-2-20261009.wav", "sha256": "9d6af912ed5ac664d8920b175a3479aaef9069006df1d3c784b561248b918ce9", "float_sha256": "b6f698bdc4aae7a0469f5bb4868315ec320ddf4bb69d0cfb265eae803f8f369d", "generation_seconds": 0.0, "audio_seconds": 5.72, "rtf": 0.0, "mlx_peak_bytes": 4637517236, "rms": 0.07540756300671848, "peak": 0.549577534198761, "clipped_fraction": 0.0, "prompt": 2, "lang": "en", "text": "Please check the appointment on Thursday at three thirty. Bring the blue folder, your keys, and a bottle of water.", "seed": 20261009 } }, { "file": "fp32-decoder-2-42.wav", "sha256": "76440a13ca19e5e7fbcdc9bfc03dd9200662996a6cd83b0baf60560b371e79fe", "float_sha256": "5184b5b300e8e7c27816dc720c462d44863ead408fe4fcb2f2c4e09649f0cfee", "generation_seconds": 2.4333328750071814, "audio_seconds": 5.32, "rtf": 0.45739339755774083, "mlx_peak_bytes": 4583268292, "rms": 0.07560389578359764, "peak": 0.5671860575675964, "clipped_fraction": 0.0, "prompt": 2, "lang": "en", "text": "Please check the appointment on Thursday at three thirty. Bring the blue folder, your keys, and a bottle of water.", "seed": 42, "decoder_seed": 9182, "speech_tokens": 133, "source_token_sha256": "13365eea735200dba4ce5afe9e981a1dd747ae44e67036cbe8704b23cc68834c", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 5.960464477539063e-08, "self_repeat_relative_rms_delta": 4.720770974131483e-08, "self_repeat_control": { "file": "fp32-decoder-self-2-42.wav", "sha256": "4a864b053ee402b34dce44d0bbc7b8446bda9a289c96e36427fac4628fa357d6", "float_sha256": "1f70c833f04bd50841a1cf736db3e90e915b988f24dfb816ba7e873dcd19028f", "generation_seconds": 0.0, "audio_seconds": 5.32, "rtf": 0.0, "mlx_peak_bytes": 4583268292, "rms": 0.07560389578680088, "peak": 0.5671860575675964, "clipped_fraction": 0.0, "prompt": 2, "lang": "en", "text": "Please check the appointment on Thursday at three thirty. Bring the blue folder, your keys, and a bottle of water.", "seed": 42 } }, { "file": "fp32-decoder-3-20261009.wav", "sha256": "afe2c844c93591c18c89ebe5f7198665d9f48d14fefcbd9634896e7b87cc7dec", "float_sha256": "af6f6033b118e9c1b23c94e074ad73375a809ab980500c13d15449a56748204d", "generation_seconds": 2.8357523750019027, "audio_seconds": 6.52, "rtf": 0.4349313458591876, "mlx_peak_bytes": 4748816788, "rms": 0.09044293349782931, "peak": 0.6633583903312683, "clipped_fraction": 0.0, "prompt": 3, "lang": "es", "text": "Este es el checkpoint de Chatterbox Turbo de oMLX. Fue cuantizado a partir del modelo original fp16.", "seed": 20261009, "decoder_seed": 9182, "speech_tokens": 163, "source_token_sha256": "912cde02e0ae66e34fefdca7455d29fe0a3a88094574e2fea2f3af2d35252004", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 8.940696716308594e-08, "self_repeat_relative_rms_delta": 4.7296964567067334e-08, "self_repeat_control": { "file": "fp32-decoder-self-3-20261009.wav", "sha256": "66cbb68730b0580d8a1e151ac3f561367305214e68813887f89410c32148ade5", "float_sha256": "fdc934f224408769a0b39eb557cda69bba1ffc362cecceee1b889e157c2b5fc1", "generation_seconds": 0.0, "audio_seconds": 6.52, "rtf": 0.0, "mlx_peak_bytes": 4748931476, "rms": 0.09044293350856547, "peak": 0.6633583903312683, "clipped_fraction": 0.0, "prompt": 3, "lang": "es", "text": "Este es el checkpoint de Chatterbox Turbo de oMLX. Fue cuantizado a partir del modelo original fp16.", "seed": 20261009 } }, { "file": "fp32-decoder-3-42.wav", "sha256": "153a94873170b4836dbcae51473a9c0a88ca4ae25b577ec584619a8bc5c1e612", "float_sha256": "8f408528c5cf332af2f2a32007ac476bf258da0ca7158951b072631323a9a6aa", "generation_seconds": 2.717131874989718, "audio_seconds": 6.4, "rtf": 0.42455185546714347, "mlx_peak_bytes": 4729171940, "rms": 0.08401561327205886, "peak": 0.6785795092582703, "clipped_fraction": 0.0, "prompt": 3, "lang": "es", "text": "Este es el checkpoint de Chatterbox Turbo de oMLX. Fue cuantizado a partir del modelo original fp16.", "seed": 42, "decoder_seed": 9182, "speech_tokens": 160, "source_token_sha256": "510a36cc71321f20b5a06826b4884130d4adc8618585757ffdc2e74ec5899b9e", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 8.940696716308594e-08, "self_repeat_relative_rms_delta": 4.73708752224411e-08, "self_repeat_control": { "file": "fp32-decoder-self-3-42.wav", "sha256": "d96992eda9234da25085ce8e389e91a080bd6af4ac43918b0273a1cd1480c82c", "float_sha256": "d92c35097c10a1c7f34e6ecebba67b52dd1ee4d3621d559732b306ccaf674b90", "generation_seconds": 0.0, "audio_seconds": 6.4, "rtf": 0.0, "mlx_peak_bytes": 4729171940, "rms": 0.08401561328234727, "peak": 0.6785795092582703, "clipped_fraction": 0.0, "prompt": 3, "lang": "es", "text": "Este es el checkpoint de Chatterbox Turbo de oMLX. Fue cuantizado a partir del modelo original fp16.", "seed": 42 } }, { "file": "fp32-decoder-4-20261009.wav", "sha256": "45822d21c94a95932d51693bd98d548b746707a611830d2474696fb13e29099b", "float_sha256": "3db17132ae1512962f4af9bbeb33523c9ea0e5abbc7becffc57ff6b984347c23", "generation_seconds": 2.705148291992373, "audio_seconds": 6.08, "rtf": 0.44492570591979813, "mlx_peak_bytes": 4685212460, "rms": 0.08884341007363974, "peak": 0.8404066562652588, "clipped_fraction": 0.0, "prompt": 4, "lang": "es", "text": "Hola, amigo, el checkpoint de entrenamiento está guardado y este es Chatterbox hablando localmente en la Mac.", "seed": 20261009, "decoder_seed": 9182, "speech_tokens": 152, "source_token_sha256": "bbee0e53132295f6c223b590727068ebf1b10ddc1a98d20c9889f806c232e742", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 1.1920928955078125e-07, "self_repeat_relative_rms_delta": 4.8030845078983475e-08, "self_repeat_control": { "file": "fp32-decoder-self-4-20261009.wav", "sha256": "eb79d2b14e59d52ae0063c267a9f9552273517f35c4ea03090e3a366660b9c5a", "float_sha256": "9a3da4eaa7f4a59542fa763983466b2d432c5a0a4665d69821d5b3385bae38aa", "generation_seconds": 0.0, "audio_seconds": 6.08, "rtf": 0.0, "mlx_peak_bytes": 4685212460, "rms": 0.0888434100541644, "peak": 0.8404066562652588, "clipped_fraction": 0.0, "prompt": 4, "lang": "es", "text": "Hola, amigo, el checkpoint de entrenamiento está guardado y este es Chatterbox hablando localmente en la Mac.", "seed": 20261009 } }, { "file": "fp32-decoder-4-42.wav", "sha256": "5cbd0585b95acdd0e53817cecc146f9704645495e1e4126de7dedeeb2736e017", "float_sha256": "362dccfec5a444358499778765180aad8fac995df1bbc16d477ec1165616922c", "generation_seconds": 2.609158250008477, "audio_seconds": 6.16, "rtf": 0.4235646509754021, "mlx_peak_bytes": 4695256124, "rms": 0.08316910082945266, "peak": 0.7973048090934753, "clipped_fraction": 0.0, "prompt": 4, "lang": "es", "text": "Hola, amigo, el checkpoint de entrenamiento está guardado y este es Chatterbox hablando localmente en la Mac.", "seed": 42, "decoder_seed": 9182, "speech_tokens": 154, "source_token_sha256": "a1be4dc194c481b4c1dd62337f92e6bd9769fa6465784acc4cfa4f018131704e", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 8.940696716308594e-08, "self_repeat_relative_rms_delta": 4.629333361094723e-08, "self_repeat_control": { "file": "fp32-decoder-self-4-42.wav", "sha256": "51e730254ead18cb5f27f2e746787b7733b42bf0468617fb4b48409c3a5b22bc", "float_sha256": "d40c6b87efd1db2eb0ae5c8daea19d5772626e7a398f2a8cb4e7ae7c9897a072", "generation_seconds": 0.0, "audio_seconds": 6.16, "rtf": 0.0, "mlx_peak_bytes": 4695273020, "rms": 0.08316910082239877, "peak": 0.7973048090934753, "clipped_fraction": 0.0, "prompt": 4, "lang": "es", "text": "Hola, amigo, el checkpoint de entrenamiento está guardado y este es Chatterbox hablando localmente en la Mac.", "seed": 42 } }, { "file": "fp32-decoder-5-20261009.wav", "sha256": "77231dae10a4433c0d0579d4dbdd134c568379aa12d8a67e3d98d4eb36e6ade1", "float_sha256": "e04cb8586215f3a08f28102a3f63a394c7ebd2f02877d1784434f1f7f60b4ebf", "generation_seconds": 2.6004575830011163, "audio_seconds": 6.08, "rtf": 0.4277068393093941, "mlx_peak_bytes": 4685196052, "rms": 0.09924828453372422, "peak": 0.7692153453826904, "clipped_fraction": 0.0, "prompt": 5, "lang": "es", "text": "Por favor, revisa la cita del jueves a las tres y media. Trae la carpeta azul, tus llaves y una botella de agua.", "seed": 20261009, "decoder_seed": 9182, "speech_tokens": 152, "source_token_sha256": "aa5fff77bc601f519ec9ba885eb0a4918aad4ef242fd076670f4ffef9fd50847", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 1.1920928955078125e-07, "self_repeat_relative_rms_delta": 4.653313112612523e-08, "self_repeat_control": { "file": "fp32-decoder-self-5-20261009.wav", "sha256": "a7a409b044404c4c3445081b02eec9e87d43b880cbb03f3bc0062257b970d7b5", "float_sha256": "b205e50222c0539cb13e63afbecaeb2e7d57407d482ff4bad4f993d187858ca9", "generation_seconds": 0.0, "audio_seconds": 6.08, "rtf": 0.0, "mlx_peak_bytes": 4685196052, "rms": 0.09924828453574455, "peak": 0.7692153453826904, "clipped_fraction": 0.0, "prompt": 5, "lang": "es", "text": "Por favor, revisa la cita del jueves a las tres y media. Trae la carpeta azul, tus llaves y una botella de agua.", "seed": 20261009 } }, { "file": "fp32-decoder-5-42.wav", "sha256": "afea4f5f099f725926d82fd4140abbb1ee36478f7db74f7a13bff35ea0fe4ac9", "float_sha256": "af803464fce229a0fde0df60c01b294b430ef3bc2bd56a05ee3b9c7d821c068a", "generation_seconds": 3.317616167012602, "audio_seconds": 8.28, "rtf": 0.4006782810401694, "mlx_peak_bytes": 4728308772, "rms": 0.08350007324714463, "peak": 0.9702610969543457, "clipped_fraction": 0.0, "prompt": 5, "lang": "es", "text": "Por favor, revisa la cita del jueves a las tres y media. Trae la carpeta azul, tus llaves y una botella de agua.", "seed": 42, "decoder_seed": 9182, "speech_tokens": 207, "source_token_sha256": "0012d919bf529baf19b526f9ee344ad618f3963beb3bb616efac8b261443cd59", "reference_shape_matched": true, "self_repeat_bitwise_equal": false, "self_repeat_max_abs_delta": 1.1920928955078125e-07, "self_repeat_relative_rms_delta": 4.813351495158713e-08, "self_repeat_control": { "file": "fp32-decoder-self-5-42.wav", "sha256": "1551bc025890e477a7c28f3826200ab958b5f03aec6f85a0448341b46a54e3a7", "float_sha256": "209a74c71aa0b2f429f0439af54f813c989db0b630d4d9f6f222c1b76faaf5f3", "generation_seconds": 0.0, "audio_seconds": 8.28, "rtf": 0.0, "mlx_peak_bytes": 4728521764, "rms": 0.08350007322332566, "peak": 0.9702610969543457, "clipped_fraction": 0.0, "prompt": 5, "lang": "es", "text": "Por favor, revisa la cita del jueves a las tres y media. Trae la carpeta azul, tus llaves y una botella de agua.", "seed": 42 } } ], "full_process_peak_mlx_bytes": 4752991988, "process_peak_rss_bytes": 3469230080, "warm_median_seconds": 7.854540125001222, "warm_min_seconds": 7.7425289589882595, "warm_max_seconds": 8.44908179200138, "warm_median_rtf": 0.926242939269012, "warm_same_audio": false, "warm_profile_peak_mlx_bytes": 4752006708, "shared_decoder_conditionals_sha256": "209d21a9230f596ffc854ad305919a3fc33a4bdc746e60baf67239ca9a17043f", "decoder_flow_signature": "(token: mlx.core.array, token_len: mlx.core.array, prompt_token: mlx.core.array, prompt_token_len: mlx.core.array, prompt_feat: mlx.core.array, prompt_feat_len: mlx.core.array, embedding: mlx.core.array, finalize: bool, n_timesteps: Optional[int] = None, streaming: bool = False)", "decoder_method": "Deployed S3Token2Wav.__call__, finalize=True, original FP32 tokens and FP32 decoder conditionals; fixed MLX RNG seed." }