PeopleofSuperintelligence

Researcher

Sanchit Gandhi

Sanchit Gandhi is a research scientist at Mistral who builds open speech and language models. Holding a master's degree from the University of Cambridge, he worked as a machine learning research engineer for audio on Hugging Face's open-source team, where he co-authored the ESB speech benchmark in October 2022 and wrote a guide to fine-tuning Whisper in November 2022. In November 2023, he co-authored Distil-Whisper, a distilled model that is 5.8 times faster than Whisper with 51% fewer parameters. At Mistral, he co-authored Magistral, the company's first reasoning model, in June 2025, and the Voxtral audio models in July 2025. He also co-authored the Open ASR Leaderboard paper in October 2025, followed by Voxtral Realtime and Voxtral TTS in 2026.

Works
15
Citations
977
h-index
4

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. October 2022

    The ESB speech benchmark

    Holding a master's degree from the University of Cambridge, he co-authors the End-to-end Speech Benchmark (ESB) in October 2022 with Patrick von Platen and Alexander M. Rush. ESB evaluates one speech recognition system across a broad set of speech datasets.

  2. November 2022

    A Whisper fine-tuning guide

    In November 2022, he publishes a step-by-step Hugging Face blog guide to fine-tuning OpenAI's Whisper for multilingual speech recognition with Transformers. At Hugging Face he is a machine learning research engineer for audio on the open-source team, working to make speech models faster, lighter and easier to use.

  3. 2023

    Distil-Whisper and faster decoding

    He co-authors Distil-Whisper in November 2023; the distilled Whisper model is 5.8 times faster with 51% fewer parameters and performs within 1% word error rate on out-of-distribution test data. A Hugging Face blog post of his in December 2023 shows speculative decoding cutting Whisper's inference time by a factor of 2 with exactly the same outputs.

  4. June 2025

    Co-authoring Magistral

    In June 2025, he is among the authors of Magistral, Mistral's first reasoning model, trained with the company's own scalable reinforcement learning pipeline. Mistral open-sources Magistral Small under Apache 2.0.

  5. July 2025

    The Voxtral audio models

    The Voxtral paper of July 2025 lists him among its authors. Voxtral Mini and Voxtral Small are two multimodal audio chat models trained to comprehend both spoken audio and text documents.

  6. October 2025

    The Open ASR Leaderboard

    In October 2025, he co-authors the Open ASR Leaderboard paper, a reproducible benchmarking platform that compares 86 open-source and proprietary speech recognition systems across 12 datasets.

  7. February 2026–March 2026

    Voxtral Realtime and Voxtral TTS

    He co-authors Voxtral Realtime in February 2026, a natively streaming speech recognition model that performs on par with Whisper at a delay of 480ms. In March 2026, he co-authors Voxtral TTS, a multilingual text-to-speech model that achieves a 68.4% win rate over ElevenLabs Flash v2.5 in human evaluations by native speakers.

  8. As of October 2026

    Research scientist at Mistral

    He is a research scientist at Mistral, based in London, after earlier open-source work at Hugging Face.

Career

  1. Since 2025Mistral AIResearch scientist
  2. Hugging FaceMachine learning research engineer for audio

Education

  1. University of CambridgeMaster's degree

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

Most-cited works

  1. BLOOM: A 176B-Parameter Open-Access Multilingual Language Model2023 · 647 citationsWith Alexander Rush, Anna Rogers, Dragomir Radev, Kyle Lo, Margaret Mitchell and Niklas Muennighoff
  2. Distil-Whisper: Robust Knowledge Distillation via Large-Scale Pseudo Labelling2023 · 18 citationsWith Alexander Rush
  3. ESB: A Benchmark For Multi-Domain End-to-End Speech Recognition2022 · 13 citationsWith Alexander Rush
  4. Ministral 32026 · 4 citationsWith Abhinav Rastogi, Alexander H. Liu, Alexandre Sablayrolles, Amélie Héliou, Andy Ehrenberg and Andy Lo
  5. Magistral2025 · 3 citationsWith Alexander H. Liu, Alexandre Sablayrolles, Amélie Héliou, Andy Ehrenberg, Baptiste Bout and Baudouin De Monicault

From OpenAlex, as of 21 October 2026.

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Sanchit Gandhi. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  2. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes Mistral AI. Our estimates of the organizations where 550 people worked directly before joining 9 frontier AI labs across 620 career transitions, comparing the main feeder organizations per lab.
  3. Of 943 people shaping AI who worked at Google, 446 work there nowIncludes Mistral AI. Our estimates of where the 943 people shaping AI who worked at Google and Google DeepMind work now: 446 at Google and 452 across 278 other organizations.
  4. 5,000 frontier users would be 38.2% of the core labs' 13,100 reported staffIncludes Mistral AI. 5,000 frontier users would be 38.2% of the core labs' 13,100 reported staff. We test the estimate against 3,913 named model contributors and dated all-staff records across Google DeepMind, OpenAI and Anthropic, plus six additional frontier-grade labs.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Sanchit Gandhi yet. Send us a free-licensed photo.

  1. Sanchit Gandhi - Hugging Face huggingface.co. Checked 20 October 2026.
  2. ESB: A Benchmark For Multi-Domain End-to-End Speech Recognition arxiv.org. Checked 20 October 2026.
  3. Fine-Tune Whisper For Multilingual ASR with Transformers huggingface.co. Checked 20 October 2026.
  4. Welcome to the Hugging Face Audio course! huggingface.co. Checked 20 October 2026.
  5. Distil-Whisper: Robust Knowledge Distillation via Large-Scale Pseudo Labelling arxiv.org. Checked 20 October 2026.
  6. Speculative Decoding for 2x Faster Whisper Inference huggingface.co. Checked 20 October 2026.
  7. Magistral arxiv.org. Checked 20 October 2026.
  8. Voxtral arxiv.org. Checked 20 October 2026.
  9. Open ASR Leaderboard: Towards Reproducible and Transparent Multilingual and Long-Form Speech Recognition Evaluation arxiv.org. Checked 20 October 2026.
  10. Voxtral Realtime arxiv.org. Checked 20 October 2026.
  11. Voxtral TTS arxiv.org. Checked 20 October 2026.
  12. sanchit-gandhi (Sanchit Gandhi) · GitHub github.com. Checked 20 October 2026.