RunInfra is now public.See what's new
RunInfra
Own your AI. We benchmark GPUs, optimize kernels, and deploy open-source models as production APIs.
Start building
Transcribe, diarize, redact on open Whisper.
Long-form audio to searchable transcripts. Speaker labels, PII redaction, exportable on your stack.
Meetings
Calls
Redaction
Search
What you actually own
The optimization knobs, the codebase, the model choice. None of it locked away.
01
Long-form, not per-minute.
Closed transcription APIs charge per audio minute and stop at the transcript. RunInfra runs diarization, redaction, and search in the same job for one per-million-tokens unit cost.
02
PII never leaves your stack.
Phone recordings, meetings, and dictations stay on your runtime. Redaction runs before the transcript ever touches storage or search.
03
Search every call afterward.
Diarized, redacted transcripts get a hybrid full-text and semantic index. Search, audit, and export the corpus without a second vendor.
Three ways to ship long-form transcription
Most teams pick between speed and control. RunInfra keeps both in one workflow.
| What matters | RunInfraRecommendedFast path with model control and export. | Closed transcription APIsPer-minute, hosted. | DIY self-hostingFull control, heavy operations. |
|---|---|---|---|
| 01Launch | Pick model, optimize, deploy Start quickly and keep the production path open. | Call provider endpoint Fast first demo, but the runtime stays rented. | Build serving stack first Infrastructure work comes before product learning. |
| 02Model control | Bring the model ID Keep model choice and serving decisions visible. | Provider catalog You use what the provider exposes. | Your model Full control if your team maintains the runtime. |
| 03Tuning | Measured latency and GPU cost Compare serving choices before deployment. | Opaque Latency and batching stay behind the API. | Manual profiling Your team owns tuning and regressions. |
| 04Export | Managed now, export when needed Use the endpoint first and take the deploy package later. | Locked endpoint You keep calling the provider. | Already owned Export exists because you built everything yourself. |
| 05Operations | Low until you choose to own it Operate managed, then export with the same measured plan. | Low, with lock-in Less infra work, less production control. | High You own infra, failures, upgrades, and serving changes. |
| 06Security | SOC 2 Type 2 Audited controls across access, logging, and incident response. | Varies by vendor Compliance depends on the third party sitting in the audio path. | You build it Your team owns the audit trail, logging, and access controls. |
RunInfra
RecommendedFast path with model control and export.
Launch
Pick model, optimize, deploy
Start quickly and keep the production path open.
Model control
Bring the model ID
Keep model choice and serving decisions visible.
Tuning
Measured latency and GPU cost
Compare serving choices before deployment.
Export
Managed now, export when needed
Use the endpoint first and take the deploy package later.
Operations
Low until you choose to own it
Operate managed, then export with the same measured plan.
Security
SOC 2 Type 2
Audited controls across access, logging, and incident response.
Code you own. Deploy anywhere.
The full recipe ships with you. Codebase, kernels, engine config, weights. Run it anywhere.
Live
Add a custom PII entity to the redactor...
Pipeline modules5
runinfra-transcription/
audio/
evals/
Managed RunInfra
Our GPUs, per-million-tokens billing from L4 to B200.
Your infrastructure
AWS, GCP, RunPod, bare metal. Same Dockerfile, your cluster.
Local workstation
docker compose up. Full pipeline on a single GPU.
Any HF ASR model works
Every Whisper-class model on Hugging Face runs through the same recipe. Search the live catalog above. The examples below are just a starting view.
HF
What RunInfra tunes
Every stage of the pipeline, retuned per model and GPU.
Batch ASR concurrency
Continuous batching across audio shards. 120x realtime on a single L4 for offline jobs.
Whisper kernels
Mel-spectrogram fusion. FlashAttention v2 on the encoder. Same kernels for streaming and batch.
Speaker diarization
Pyannote-based speaker turns. Swap diarizers without retuning the ASR step.
PII redaction
Entity-scoped redaction for names, emails, phones, account IDs, and your custom fields.
Searchable transcript index
Hybrid full-text and semantic search across diarized, redacted segments.
Word and segment timestamps
Per-word time anchors plus speaker-segment ranges. SRT, VTT, JSON, and warehouse exports.
Try this pipeline
Edit the model, engine, or GPU inline. Send to retune the stack in the dashboard.
Common questions
Can't find what you're looking for? Get in touch
How does PII redaction work?
Named entities (person, email, phone, account ID) get masked in the transcript before it is stored or indexed. Custom entity types can be added with a regex or a small classifier. Redaction runs as a sibling step so the original audio is never required downstream.
