audio workload · 33 runs on record

Whisper Large v3

OpenAI Whisper Large v3, FP16, 30s English audio, single stream. Higher = faster than real time.

Primary metric: Real-time factor (x RT)

Reference prompts

Representative prompts for this workload. Exact prompts and harness settings still depend on the cited source for each record.

  • Prompt 1

    30-second clip — clean conference talk, English (LibriSpeech test-clean).

  • Prompt 2

    30-second clip — podcast crosstalk, 2 speakers, English.

  • Prompt 3

    30-second clip — accented English (CommonVoice subset).

Reference runtime command

A representative invocation for reproducing this workload class. Source-specific runs may use adjacent runtimes unless the record says otherwise.

shell
whisper-cli -m ggml-large-v3.bin -f sample.wav --language en --output-txt

Real-time factor = (audio duration ÷ transcription wall time). Higher = faster than real-time.

Full leaderboard

Every record for Whisper transcription.

Top 10 leaderboard

Whisper Large v3 · sorted by real-time factor

Bar chart: Top 10 devices ranked by Real-time factor for Whisper Large v3. 1. NVIDIA B200 192GB at 210 x RT. 2. NVIDIA H100 SXM5 80GB at 145 x RT. 3. NVIDIA GeForce RTX 5090 32GB at 128 x RT. 4. NVIDIA GeForce RTX 5090 32GB at 105 x RT. 5. NVIDIA GeForce RTX 5080 16GB at 95 x RT.

Value frontier, MSRP vs real-time factor

Each dot is a device. Top-left is best value (cheap + fast).

Scatter chart of MSRP versus Real-time factor across 33 devices. Devices in the upper-left region offer the best price/performance ratio. Top 5 by primary metric: B200 192GB at $39,999 delivering 210 x RT; H100 SXM5 80GB at $25,000 delivering 145 x RT; GeForce RTX 5090 32GB at $1,999 delivering 128 x RT; GeForce RTX 5090 32GB at $1,999 delivering 105 x RT; GeForce RTX 5080 16GB at $999 delivering 95 x RT.

Full leaderboard

Click any row for detailed breakdown. Click column headers to sort.

#DeviceVerifBuy
1
NVIDIA B200 192GBDatacenter GPU
NVIDIA·FP16
6
210.0xx RT
192 GB1000 W$40k——Amazon
2
NVIDIA H100 SXM5 80GBDatacenter GPU
NVIDIA·FP16
6
145.0xx RT
80 GB700 W$25k——Amazon
3
NVIDIA GeForce RTX 5090 32GBConsumer GPU
NVIDIA·FP16
6
128.0xx RT
32 GB575 W$2.0k——Amazon
#4
NVIDIA GeForce RTX 5090 32GBConsumer GPU
NVIDIA·FP16
6
105.0xx RT
32 GB575 W$2.0k——Amazon
#5
NVIDIA GeForce RTX 5080 16GBConsumer GPU
NVIDIA·FP16
6
95.0xx RT
16 GB360 W$999——Amazon
#6
AMD Instinct MI300X 192GBDatacenter GPU
AMD·FP16
6
95.0xx RT
192 GB750 W$15k——Amazon
#7
NVIDIA GeForce RTX 4080 Super 16GBConsumer GPU
NVIDIA·FP16
6
82.0xx RT
16 GB320 W$999——Amazon
#8
NVIDIA GeForce RTX 4090 24GBConsumer GPU
NVIDIA·FP16
6
78.0xx RT
24 GB450 W$1.6k——Amazon
#9
NVIDIA GeForce RTX 5090 32GBConsumer GPU
NVIDIA·FP16
6
78.0xx RT
32 GB575 W$2.0k——Amazon
#10
NVIDIA GeForce RTX 4080 Super 16GBConsumer GPU
NVIDIA·FP16
6
64.0xx RT
16 GB320 W$999——Amazon
#11
NVIDIA GeForce RTX 3090 24GBConsumer GPU
NVIDIA·FP16
6
56.0xx RT
24 GB350 W$1.5k——Amazon
#12
AMD Radeon RX 7900 XTX 24GBConsumer GPU
AMD·FP16
6
48.0xx RT
24 GB355 W$999——Amazon
#13
NVIDIA GeForce RTX 4080 Super 16GBConsumer GPU
NVIDIA·FP16
6
48.0xx RT
16 GB320 W$999——Amazon
#14
NVIDIA GeForce RTX 4070 12GBConsumer GPU
NVIDIA·FP16
6
42.0xx RT
12 GB200 W$599——Amazon
#15
Apple M4 Max (40c GPU, 128GB)Apple Silicon
Apple·FP16
6
32.0xx RT
128 GB70 W$4.7k——Amazon
#16
NVIDIA GeForce RTX 3090 24GBConsumer GPU
NVIDIA·FP16
6
32.0xx RT
24 GB350 W$1.5k——Amazon
#17
NVIDIA GeForce RTX 4060 Ti 16GBConsumer GPU
NVIDIA·FP16
6
30.0xx RT
16 GB165 W$499——Amazon
#18
Intel Arc B580 12GBConsumer GPU
Intel·INT8
6
28.0xx RT
12 GB190 W$249——Amazon
#19
Apple M3 Max (40c GPU, 64GB)Apple Silicon
Apple·FP16
6
26.0xx RT
64 GB65 W$3.5k——Amazon
#20
Apple M4 Pro (20c GPU, 48GB)Apple Silicon
Apple·FP16
6
22.0xx RT
48 GB50 W$2.5k——Amazon
#21
AMD Radeon RX 7900 XTX 24GBConsumer GPU
AMD·FP16
6
22.0xx RT
24 GB355 W$999——Amazon
#22
Intel Arc B580 12GBConsumer GPU
Intel·FP16
6
18.0xx RT
12 GB190 W$249——Amazon
#23
Apple Mac mini M4 Pro 48GBApple Silicon
Apple·FP16
6
18.0xx RT
48 GB35 W$2.0k——Amazon
#24
Apple M4 Max (40c GPU, 128GB)Apple Silicon
Apple·FP16
6
12.5xx RT
128 GB65 W$4.7k——Amazon
#25
NVIDIA Jetson AGX Orin 64GBEdge
NVIDIA·FP16
6
11.0xx RT
64 GB60 W$2.0k——Amazon
#26
AMD Ryzen AI Max+ 395 (Strix Halo, 96GB)NPU
AMD·INT8
6
8.0xx RT
96 GB120 W$2.2k——Amazon
#27
NVIDIA Jetson Orin Nano Super 8GBEdge
NVIDIA·FP16
6
4.5xx RT
8 GB25 W$249——Amazon
#28
Intel Core Ultra 7 258V (Lunar Lake)NPU
Intel·INT8
6
4.2xx RT
, 17 W$1.3k——Amazon
#29
NVIDIA Jetson AGX Orin 64GBEdge
NVIDIA·FP16
6
4.2xx RT
64 GB60 W$2.0k——Amazon
#30
NVIDIA Jetson Orin Nano 8GBEdge
NVIDIA·FP16
6
2.5xx RT
8 GB15 W$499——Amazon
#31
Qualcomm Snapdragon X Elite (X1E-84-100)NPU
Qualcomm·INT8
6
2.1xx RT
, 23 W$1.2k——Amazon
#32
Raspberry Pi AI Hat+ (Hailo-8L 13 TOPS)Edge
Raspberry Pi·INT8
6
1.2xx RT
, 8 W$70——Amazon
#33
Raspberry Pi 5 + AI HAT+ (Hailo-8)Edge
Raspberry Pi·INT8
6
0.6xx RT
8 GB12 W$150——Amazon
Sorted by Real-time factor (high → low)

Cite this benchmark

Use this in your paper, blog post, or comparison table.

BibTeX
@misc{myaihardware_whisper-large-v3_2026,
  title  = {MyAI Bench: Whisper Large v3},
  author = {{MyAIHardware Contributors}},
  year   = {2026},
  url    = {https://www.myaihardware.com/benchmarks/workload/whisper-large-v3},
  note   = {Version 1.3, accessed 2026-08-27}
}
APA
MyAIHardware Contributors. (2026). MyAI Bench: Whisper Large v3. MyAIHardware. Retrieved 2026-08-27, from https://www.myaihardware.com/benchmarks/workload/whisper-large-v3
MLA
MyAIHardware Contributors. "MyAI Bench: Whisper Large v3." MyAIHardware, 2026, https://www.myaihardware.com/benchmarks/workload/whisper-large-v3. Accessed 2026-08-27.
Plain text
MyAI Bench, Whisper Large v3. MyAIHardware Contributors, 2026. Version 1.3. https://www.myaihardware.com/benchmarks/workload/whisper-large-v3 (accessed 2026-08-27).