SAVRN Model Hub · Comparisons
Qwen3-ASR-1.7B vs Voxtral-Mini-4B-Realtime-2602
Qwen3-ASR-1.7B has 2.3B parameters and Voxtral-Mini-4B-Realtime-2602 has 4.4B parameters; both are released under Apache License 2.0; at 16-bit, Qwen3-ASR-1.7B needs about 5.6 GB (1x MI300X from $1.85 an hour) and Voxtral-Mini-4B-Realtime-2602 about 10.6 GB (1x MI300X from $1.85 an hour).
| Field | Qwen3-ASR-1.7B Qwen/Qwen3-ASR-1.7B | Voxtral-Mini-4B-Realtime-2602 mistralai/Voxtral-Mini-4B-Realtime-2602 |
|---|---|---|
| Publisher | Qwen | Mistral AI_ |
| Task | Speech recognition | Speech recognition |
| Modality | Audio | Audio |
| Parameters, as reported | 2.3B parameters | 4.4B parameters |
| Architecture | Qwen3ASRForConditionalGeneration | VoxtralRealtimeForConditionalGeneration |
| Library | Not stated | vllm |
| Context length | Not stated | 131,072 tokens |
| Repository size | 4.7 GB | 17.7 GB |
| Artifact formats | safetensors | safetensors |
| License | apache-2.0 | apache-2.0 |
| Access | Open weights, no gate | Open weights, no gate |
| Memory at 16-bit (weights and margin) | 5.6 GB | 10.6 GB |
| Cheapest GPUs at 16-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Memory at 4-bit (weights and margin) | 1.4 GB | 2.7 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | 7278e1e70fe2 | 2769294da956 |
| Downloads reported by the hub | 2.4M | 2M |
| Last observed | 2026-09-18 | 2026-09-18 |
An evaluation row appears only where at least two of these models report the same benchmark with the same stated configuration, metric, unit and setup. Different evaluators stay named in each cell. Values are shown as reported: no unit conversion, no ranking.
Other Reported Results
These results are listed for each model on its own, because the conditions needed to compare them are not stated or do not match. Two results that leave a condition blank are not assumed to share it.
Qwen3-ASR-1.7B
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| hf-audio/open-asr-leaderboard | Task ami_werMetric ami_werComparison conditions not established | 10.56 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-28 |
| hf-audio/open-asr-leaderboard | Task earnings22_werMetric earnings22_werComparison conditions not established | 10.25 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-28 |
| hf-audio/open-asr-leaderboard | Task gigaspeech_werMetric gigaspeech_werComparison conditions not established | 8.74 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-28 |
| hf-audio/open-asr-leaderboard | Task librispeech_clean_werMetric librispeech_clean_werComparison conditions not established | 1.63 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-28 |
| hf-audio/open-asr-leaderboard | Task librispeech_other_werMetric librispeech_other_werComparison conditions not established | 3.4 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-28 |
| hf-audio/open-asr-leaderboard | Task mean_werMetric mean_werComparison conditions not established | 5.76 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-28 |
| hf-audio/open-asr-leaderboard | Task rtfxMetric rtfxComparison conditions not established | 147.93 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-28 |
| hf-audio/open-asr-leaderboard | Task spgispeech_werMetric spgispeech_werComparison conditions not established | 2.84 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-28 |
| hf-audio/open-asr-leaderboard | Task tedlium_werMetric tedlium_werComparison conditions not established | 2.28 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-28 |
| hf-audio/open-asr-leaderboard | Task voxpopuli_werMetric voxpopuli_werComparison conditions not established | 6.35 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-28 |
Voxtral-Mini-4B-Realtime-2602
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| hf-audio/open-asr-leaderboard | Task ami_werMetric ami_werComparison conditions not established | 17.07 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-21 |
| hf-audio/open-asr-leaderboard | Task earnings22_werMetric earnings22_werComparison conditions not established | 11.84 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-21 |
| hf-audio/open-asr-leaderboard | Task gigaspeech_werMetric gigaspeech_werComparison conditions not established | 10.38 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-21 |
| hf-audio/open-asr-leaderboard | Task librispeech_clean_werMetric librispeech_clean_werComparison conditions not established | 2.08 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-21 |
| hf-audio/open-asr-leaderboard | Task librispeech_other_werMetric librispeech_other_werComparison conditions not established | 5.52 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-21 |
| hf-audio/open-asr-leaderboard | Task mean_werMetric mean_werComparison conditions not established | 7.68 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-21 |
| hf-audio/open-asr-leaderboard | Task rtfxMetric rtfxComparison conditions not established | 93.32 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-21 |
| hf-audio/open-asr-leaderboard | Task spgispeech_werMetric spgispeech_werComparison conditions not established | 2.42 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-21 |
| hf-audio/open-asr-leaderboard | Task tedlium_werMetric tedlium_werComparison conditions not established | 3.79 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-21 |
| hf-audio/open-asr-leaderboard | Task voxpopuli_werMetric voxpopuli_werComparison conditions not established | 8.34 | open-asr-leaderboard Reported by a third party |
Evaluated revision not stated | 2026-01-21 |
SAVRN's Notes on Qwen3-ASR-1.7B
The name says 1.7B, the weight files hold 2.3B parameters, so size by the files. It turns speech into text and identifies which of 52 languages and dialects it hears; a 0.6B sibling shares its Qwen3-Omni foundation, and a separate 0.6B forced aligner handles timestamps. At 16-bit the weights are 4.7 GB and the run needs 5.6 GB, which leaves most of a 192 GB MI300X at $1.85 an hour empty, so scale by instances per GPU, not by GPUs.
Apache 2.0 permits commercial use, modification and redistribution provided the notices stay and changes are stated. Before you commit: no context length is published, no library is named, and the only artifact is safetensors, so 8-bit at 2.8 GB or 4-bit at 1.4 GB is your own quantization work. Released January 28, 2026 and described in arXiv:2601.21337; read the paper before planning around the 52 languages.
SAVRN's Notes on Voxtral-Mini-4B-Realtime-2602
Live transcription is the job here: 4.4B parameters from Mistral AI that take audio in and stream text out, with a 131,072-token context the publisher says covers about three hours under vLLM's defaults. At 16-bit it needs 10.6 GB of memory, 8.9 GB of that weights; 5.3 GB at 8-bit, 2.7 GB at 4-bit. The cheapest setup we price, one MI300X with 192 GB at $1.85 per hour on demand, sits nearly empty under one instance, so plan for many streams per accelerator.
Apache 2.0 asks only that you keep the notices and state significant changes, patent grant included, so commercial use is simple. Two checks before committing: it is derived from Ministral-3-3B-Base-2512, and the configuration lists an 8,192-token sliding window beside the long context, so test with the recording lengths you actually handle. No Index host prices it per token yet, so plan on the hourly card cost.
Questions
Which is larger, Qwen3-ASR-1.7B or Voxtral-Mini-4B-Realtime-2602?
Voxtral-Mini-4B-Realtime-2602 (4.4B parameters) is larger than Qwen3-ASR-1.7B (2.3B parameters), by the parameter counts their publishers report.
Which is cheaper to run, Qwen3-ASR-1.7B or Voxtral-Mini-4B-Realtime-2602?
At 4-bit, Qwen3-ASR-1.7B fits on 1x MI300X from $1.85 an hour and Voxtral-Mini-4B-Realtime-2602 on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use Qwen3-ASR-1.7B commercially?
Yes. Qwen3-ASR-1.7B is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
Can I use Voxtral-Mini-4B-Realtime-2602 commercially?
Yes. Voxtral-Mini-4B-Realtime-2602 is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.