WhisperKit (Argmax) Review

OSS

On-device Apple Silicon STT (Whisper)

Among the 43 speech-to-text tools we track, WhisperKit (Argmax) has the 12th-widest language coverage - a fit for multilingual and localization projects.

From $1,330/mo

Facts verified Jul 20, 2026Website →

What we know about WhisperKit (Argmax)

WhisperKit (Argmax) sits in the speech-to-text apis category, where it is on-device Apple Silicon STT (Whisper). We keep this WhisperKit (Argmax) profile grounded in primary sources, each fact dated to when we last confirmed it.

On pricing, WhisperKit (Argmax) starts at $1,330 per month for its entry tier. That is the sticker rate: real production cost usually runs higher once you add a language model, a voice provider, and telephony minutes.

On capabilities, WhisperKit (Argmax) covers language auto-detection, word-level timestamps, custom vocabulary / keyterm boosting, speech translation, and self-host / on-prem option, and does not offer entity detection, sentiment analysis, and summarization endpoint. Each of those is verified against WhisperKit (Argmax)'s own docs or dashboard, not marketing copy.

WhisperKit (Argmax) ranks the 12th-widest language coverage of the 43 speech-to-text tools we track, so where it lands for you depends on which of those matters more.

In total we track 26 verified facts for WhisperKit (Argmax) today, each linking the primary source it came from so you can check our work - and vendor claims we have not measured ourselves are labeled as such on the WhisperKit (Argmax) fact sheet below.

Reviewed by vsref Editorialfacts verified Jul 20, 2026Methodology →

Fact sheet

Every row independently verified

Pricing

Pricing facts
Pricing model	subscription	Jul 20
Cheapest paid plan	$1,330	Jul 20
Free tier quota	Unlimited - free OSS, on-device	Jul 20
Published volume discounts	Enterprise discounts; annual rate lower	Jul 20
Concurrency on base plan	Unlimited - runs on-device	Jul 20
Extra concurrency cost	None - on-device	Jul 20

Capabilities

Capabilities facts
Languages supported	99	Jul 20
Language auto-detection	✓ Yes	Jul 20
Speaker diarization	✗ Not available	Jul 20
PII redaction	✗ Not available	Jul 20
Word-level timestamps	✓ Yes	Jul 20
Custom vocabulary / keyterm boosting	✓ Yes	Jul 20
Entity detection	✗ No	Jul 20
Sentiment analysis	✗ No	Jul 20
Summarization endpoint	✗ No	Jul 20
Speech translation	✓ Yes	Jul 20
Priced audio-intelligence add-ons	VAD; SpeakerKit diarization, TTSKit TTS	Jul 20

Compliance & trust

Compliance & trust facts
Self-host / on-prem option	✓ Yes	Jul 20
Model weights license	MIT	Jul 20

Build experience

Build experience facts
Official SDKs	Swift (Swift Package Manager) + CLI; Pro Kotlin	Jul 20
Websocket streaming API	✗ No	Jul 20

Commercial

Commercial facts
Model size (parameters)	tiny 39M / base 74M / small 244M / large-v3 1.5B	Jul 20
Hardware to self-host	Apple Silicon (M/A-series), Core ML + ANE	Jul 20
Hosted API available	No - on-device SDK only	Jul 20
Project maintenance status	Active - v1.0.0 released 2026-05	Jul 20
GitHub stars	6,279	Jul 20

Considering a switch? Best WhisperKit (Argmax) alternatives →

Voice agents in this stack

The engine is one layer: a voice agent hears through its transcription engine, but orchestration, telephony, and turn-taking come from the agent platform. Compare the platforms builders pair WhisperKit (Argmax) with: Pipecat, OpenAI Realtime API, Twilio ConversationRelay, or the full voice-agent comparison →

Want it done for you?

WhisperKit (Argmax) gets you a transcript; the bot that joins the call, labels speakers, and writes the summary is still your build. If that is more pipeline than you want to own, AI meeting notetakers do the whole job end to end: Otter.ai, Fireflies.ai, Fathom, or the full notetaker comparison.