# Nemotron 3.5 Transcribe 30B A4B Benchmark Scores & Performance

> Nemotron 3.5 Transcribe 30B A4B has NVIDIA launch evaluation results. Voice evidence is display-only and does not produce a text-model score.

Canonical page: https://benchlm.ai/models/nemotron-3-5-transcribe-30b-a4b

Last updated: October 29, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | NVIDIA |
| Source Type | Open Weight |
| Reasoning Type | Reasoning |
| Context Window | 64K |
| Official model card | [NVIDIA model card](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Transcribe-30B-A4B/blob/a99e33f698257b8b69ca7834f19c73582eb1fa7e/README.md) |
| Overall Score | Not scored (voice evidence only) |
| Overall Rank | Unranked |

## Family & Coverage

- Family: Nemotron 3.5 Transcribe 30B A4B
- Variant: base
- Benchmarks covered: 0 of 724
- Coverage note: Speech and voice results appear below; no weighted text-model benchmark rows are stored.

## Speech and voice evidence

[Nemotron 3.5 Transcribe launch evaluation](/voice-benchmarks/nemotron-3-5-transcribe-launch-evaluation)

NVIDIA ran every result itself. Transcription, translation and language identification use reasoning off at temperature 0; audio understanding and text reasoning use reasoning on at temperature 0.6, top-p 0.95, with a 16,384-token reasoning budget. The English Open ASR mean is NVIDIA’s equal-weight mean of nine sets, including Monsoon English and URGENT24, so compare it only with runs over the same set list. The multilingual board uses WER throughout, while standalone FLEURS uses CER for Chinese, Japanese, Korean, Thai and Vietnamese, so those numbers are not directly comparable. The card’s per-language FLEURS, CoVoST2, language-identification and translation tables are summarised, not copied in full. The “next best” comparisons in NVIDIA’s summary tables are not carried. These results do not enter text-model or voice-agent rankings.

| Metric | Result |
| --- | --- |
| English Open ASR Leaderboard mean WER (9 sets) | 4.71% |
| Long-form Open ASR mean WER | 10.98% |
| Multilingual Open ASR Leaderboard WER (7 languages) | 4.62% |
| Multi-speaker mean cpWER (AMI, CALLHOME, Mixer 6, NOTSOFAR-1) | 15.80% |
| MMAU (full test) | 73.40% |
| Active parameters | 4.3B of 34B |

NVIDIA Hugging Face model card retrieved October 29, 2026; NVIDIA captured the results October 12, 2026. Provider-reported; error rates are better lower. Voice results are display-only.

- [Evaluation source](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Transcribe-30B-A4B)

Released October 28, 2026 under the OpenMDW-1.1 license on Hugging Face, NGC, build.nvidia.com and NeMo Speech, in BF16 with FP8 and NVFP4 variants. It runs on one H100 80 GB with vLLM or Transformers, handles recordings up to one hour within a 65,536-token context, and supports multi-token-prediction speculative decoding plus an optional DFlash2 draft model. Reasoning is switchable in the chat template. NVIDIA describes it as offline: internal chunking does not make it a real-time streaming service.

## Other NVIDIA Models

- [Nemotron 3.5 Super VL 120B A12B](/models/nemotron-3-5-super-vl-120b-a12b) - Score: 49.03
- [Nemotron 3 Ultra](/models/nemotron-3-ultra) - Score: 47.37
- [Nemotron 3.5 Bolt 74B-A9B](/models/nemotron-3-5-bolt-74b-a9b) - Score: 42.8
- [Nemotron 3 Nano Omni 30B A3B](/models/nemotron-3-nano-omni-30b-a3b) - Score: 33.93
- [Nemotron 3.5 Lightning 30B A3B NVFP4](/models/nemotron-3-5-lightning-30b-a3b-nvfp4) - Score: 26.37
- [Nemotron-4 15B](/models/nemotron-4-15b) - Score: 9.89
- [Nemotron Ultra 253B](/models/nemotron-ultra-253b) - Score: not computed
- [Nemotron 3.5 Bolt 74B-A9B BF16](/models/nemotron-3-5-bolt-74b-a9b-bf16) - Score: not computed
- [Nemotron 3.5 Super VL 120B A12B NVFP4](/models/nemotron-3-5-super-vl-120b-a12b-nvfp4) - Score: not computed
- [Nemotron 3 Super 100B](/models/nemotron-3-super-100b) - Score: not computed
- [Nemotron 3 Nano 30B](/models/nemotron-3-nano-30b) - Score: not computed
- [Nemotron 3.5 Bolt E4B BF16](/models/nemotron-3-5-bolt-e4b-bf16) - Score: not computed
- [Nemotron 3.5 Bolt E4B](/models/nemotron-3-5-bolt-e4b) - Score: not computed
- [Nemotron 3 Super 120B A12B](/models/nemotron-3-super-120b-a12b) - Score: not computed
- [Cosmos3-Edge](/models/cosmos3-edge) - Score: not computed
- [Nemotron 3.5 ASR Streaming 0.6B](/models/nemotron-3-5-asr-streaming-0-6b) - Score: not computed
- [Cosmos3-Nano](/models/cosmos3-nano) - Score: not computed
- [Cosmos3-Super](/models/cosmos3-super) - Score: not computed
- [Parakeet TDT 0.6B v3](/models/parakeet-tdt-0-6b-v3) - Score: not computed
- [Audio Flamingo 3 7B](/models/audio-flamingo-3-7b) - Score: not computed
- [Canary 180M Flash](/models/canary-180m-flash) - Score: not computed
- [Parakeet CTC 1.1B](/models/parakeet-ctc-1-1b) - Score: not computed
- [Llama-3.1-Nemotron-70B-Reward (JudgeBench)](/models/native-judgebench-llama-3-1-nemotron-70b-reward) - Score: not computed
- [Llama-3.3-Nemotron-70B-Reward (JudgeBench)](/models/native-judgebench-llama-3-3-nemotron-70b-reward) - Score: not computed
- [Llama-3.3-Nemotron-70B-Reward-Multilingual (JudgeBench)](/models/native-judgebench-llama-3-3-nemotron-70b-reward-multilingual) - Score: not computed
- [Llama-3.3-Nemotron-70B-Reward-Principle (JudgeBench)](/models/native-judgebench-llama-3-3-nemotron-70b-reward-principle) - Score: not computed
- [Llama-3.3-Nemotron-Super-49B-GenRM (JudgeBench)](/models/native-judgebench-llama-3-3-nemotron-super-49b-genrm) - Score: not computed
- [Llama-3.3-Nemotron-Super-49B-GenRM + voting@32 (JudgeBench)](/models/native-judgebench-llama-3-3-nemotron-super-49b-genrm-voting-32) - Score: not computed
- [Llama-3.3-Nemotron-Super-49B-GenRM-Multilingual (JudgeBench)](/models/native-judgebench-llama-3-3-nemotron-super-49b-genrm-multilingual) - Score: not computed
- [Llama-3.3-Nemotron-Super-49B-GenRM-Multilingual + voting@32 (JudgeBench)](/models/native-judgebench-llama-3-3-nemotron-super-49b-genrm-multilingual-voting-32) - Score: not computed
- [Qwen-2.5-Nemotron-32B-Reward (JudgeBench)](/models/native-judgebench-qwen-2-5-nemotron-32b-reward) - Score: not computed
- [Qwen-3-Nemotron-32B-Reward (JudgeBench)](/models/native-judgebench-qwen-3-nemotron-32b-reward) - Score: not computed
- [Qwen3-Nemotron-32B-GenRM-Principle (JudgeBench)](/models/native-judgebench-qwen3-nemotron-32b-genrm-principle) - Score: not computed