RTX PRO 6000 Blackwell Server Edition · 96 GB
IndexTTS-2.5
Open a report for complete measurements.
- Task
- Speech synthesis · Not recorded → Audio
- Workload
- Not recorded → Audio · tts_prompts · Test duration 9 s
- Batch size
- Batch 1
- Format · Engine
- Native · indextts · PyTorch 2.8.0+cu128
- Unit
- audio-sec/sec
- Mean throughput
- 4.2audio-sec/sec
- Observed GPU occupancy
- 6.5 GiB
- P95 latency
- Not recorded
- Reports combined
- 1
Launch this model
This engine does not yet have an executable serving recipe.
Individual tests
Not recorded → Audio · Speech synthesis · Native · indextts · tts · audio-sec/sec
Sortable table
| Test date | Throughput | Concurrency / batch | P95 (s) | TTFT (ms) | Observed occupancy (GiB) | GPU power (W) | Decode (tok/s) | Uncached prefill (tok/s) | Parameters | View report |
|---|---|---|---|---|---|---|---|---|---|---|
| 10/19/2026, 8:09:44 AM | 4.24 audio-sec/sec | 1 | — | — | 6.48 | 221.73 | — | Uncached prefill unavailableView report | Parameters ↗ | a46ec2c4 ↗ |
Recorded test parameters
Configuration 1 · Parameters
- model
- IndexTeam/IndexTTS-2.5
- engine
- indextts
- quant
- Native
- method
- audio-native-v1
- task
- tts
- dataset
- tts_prompts
- context
- tts
- input_tokens
- 0
- output_tokens
- 0
- duration
- 9
- concurrency
- 1
- unit
- audio-sec/sec
- software
- { "torch": "2.8.0+cu128", "transformers": "4.52.1", "torchaudio": "2.8.0+cu128", "indextts": "2.0.0", "OMP_NUM_THREADS": "8", "MKL_NUM_THREADS": "8", "OPENBLAS_NUM_THREADS": "8", "NUMEXPR_NUM_THREADS": "8" }
- engine_provenance
- —
- kernel_settings
- —
- precision
- —
- resolution
- —
- performance_task
- —
- performance_scope
- —
- batch_size
- 1
- dimensions
- —
- weight_quantization
- —
- max_sequence_length
- —
- model_scope
- —
- performance_dataset
- —
- workload_mode
- —
- graph_mode
- —
- model_revision
- c39ce5ba981572cb187443877ff559dfb246ce63
- client_version
- python-native-bench-1.1.3