| [audio.cpp] C++/GGML based Voxtral Mini 4B Realtime impl. 16.39x (0.0610 RTF) realtime offline, 170ms TTFT streaming on RTX 5090 | audio-cpp | open | 0 | disc |
| Unable to reproduce benchmarks | abhinavkulkarni | open | 0 | disc |
| Feature request: distilled per-language "Voxtral Micro" models for edge/on-device use Context | MaxOnAMission | open | 0 | disc |
| Add localvoxtral (macOS dictation app) to Community Contributions | T0mSIlver | open | 0 | disc |
| What is the expected ratio of GPU memory and KV cache tokens for this model? | h01-10w | open | 0 | disc |
| Training dataset size | ojo | open | 0 | disc |
| Streaming output before sending all audio | jxadro | closed | 0 | disc |
| Telephony ASR | deeksha5 | open | 0 | disc |
| model support per CrispASR — pure C++ inference with GGUF quantisation | cstr | open | 0 | disc |
| timestamp | Sh1man | open | 0 | disc |
| vllm crashing on 0.19.0 | evilperson068 | open | 0 | disc |
| Add Open ASR Leaderboard evaluation results | SaylorTwift | open | 0 | disc |
| Questions about training techniques | funnyice | open | 0 | disc |
| Voxtral Transformers path: chat template fails ("Can't compile non template nodes") and fallback output is garbled | PierrunoYT | open | 0 | disc |
| Fix tekken.json link in README.md | wilson92 | merged | 0 | disc |
| Upload top_20_wavs.zip | austinv5040 | open | 0 | disc |
| Recommendations | liuyt6515 | open | 0 | disc |
| سلام | noor18298 | open | 0 | disc |
| Dont forget to do "audio.resample(processor.feature_extractor.sampling_rate)" if adapt streaming example to your needs | trevol | open | 0 | disc |
| AttributeError when using tf v5 | P1atinum | closed | 0 | disc |
| punctuation marks | liuyt6515 | open | 0 | disc |
| i want to adapt this model for low resource african language for that i have collected the data | teeofftechnologies | open | 0 | disc |
| Update README.md | sanchit-gandhi | merged | 0 | disc |
| Optimize GPU KV cache memory usage | oleslav | closed | 0 | disc |
| Can you train indic (indian) set of languages? | trysem | open | 0 | disc |
| trfms-integration | eustlb | merged | 0 | disc |
| How to add another language ? | TheRealTancrede | open | 0 | disc |
| Voxtral Mini Realtime deciding to translate on its own ? | TheRealTancrede | closed | 0 | disc |
| Update README.md | sanchit-gandhi | merged | 0 | disc |
| Hints for language | oleslav | closed | 0 | disc |
| do not need config.json ? | Xiakj | open | 0 | disc |
| [Transformers Integration] Understanding Voxtral Realtime architecture for porting | Seungyoun | open | 0 | disc |
| Code 101101 | ifynach | closed | 0 | disc |
| AssertionError: rope can only used in combination with a sliding window | andres-llms | open | 0 | disc |
| Running on vLLM 0.15.x + Blackwell CUDA 13 | avblex | open | 0 | disc |
| diarization support? | evewashere | open | 0 | disc |
| [README.md] Specify multiples of frame-rate | sanchit-gandhi | merged | 0 | disc |
| Reducing time-to-first-token below 1s for conversational use? | jloganolson | open | 0 | disc |
| How to make sure transcription_delay_ms is changed when serving with vLLM? | goncalomcorreia | open | 0 | disc |
| Error when running on AMD with Docker | wetchoc | open | 0 | disc |
| Demo Broken | DaBrown95 | closed | 0 | disc |
| Add automatic speech recognition tag | nielsr | merged | 0 | disc |
| fix vllm parameter's name | sebag90 | merged | 0 | disc |
| Problem with super long audio | mrigankr | open | 0 | disc |
| vLLM's blog on streaming input link | jloganolson | closed | 0 | disc |
| Installation Video and Testing - Step by Step | fahdmirzac | open | 0 | disc |
| No transcription_delay param in either tekken or params | mrigankr | open | 0 | disc |
| Update README.md | iliasslasri | merged | 0 | disc |