BerriAI/litellm - GitHub vs TurboScribe
Side-by-side comparison · Updated October 2026
| Description | LiteLLM is an AI gateway and Python SDK from Berrie AI Incorporated, published in the BerriAI GitHub repository. The SDK provides a common interface for model calls inside Python applications. The proxy gateway centralizes access for a team, with virtual keys, model routing, spend tracking, budgets and an administration interface. The official documentation lists support for more than 100 model providers. Supported endpoints and features vary by integration, so verify your model’s streaming, tool-calling, image, audio or embedding requirements. The router supports retries, fallbacks and load balancing; observability integrations can send request data to tools such as Langfuse, LangSmith and OpenTelemetry. LiteLLM also provides an MCP gateway. It can connect upstream servers using Streamable HTTP, SSE or stdio, expose tools through a fixed gateway endpoint, and scope access by key, team or organization. This requires configuring the upstream servers and authentication; the gateway does not automatically grant access to third-party tools. Agent-to-agent integrations are documented separately. The open-source offering has no software license fee for self-hosting. Code outside the enterprise directory is MIT-licensed, while enterprise code has separate terms. Enterprise pricing is quoted by annual gateway request capacity, deployment architecture and support needs, rather than a per-token license charge. Model-provider charges and infrastructure costs still apply. Enterprise adds controls and support such as SSO, SCIM, audit logs and service-level agreements. Compare New API for another self-hosted gateway with provider-channel management and usage accounting. Evaluate a representative workload, inspect request logging and secret handling, test budget and failure behavior, and decide whether SDK integration or a shared gateway best fits your application. | TurboScribe turns audio and video recordings into editable text, documents and captions. It is built around file transcription: upload a recording, choose its language and transcription mode, optionally label speakers, then review and export the result. The service says it uses Whisper, supports transcription in more than 98 languages and offers transcript or subtitle translation to more than 134 languages. Speaker labels and fluent text are useful starting points, but names, numbers, quotations and changes of speaker still need checking against the recording. The free tier allows three files per day, each up to 30 minutes, with one upload at a time and lower processing priority. Unlimited costs $20 billed monthly or $120 billed yearly for one person. The advertised $10-per-month annual rate is the equivalent of that $120 yearly payment. Unlimited supports files up to 10 hours or 5 GB, up to 50 simultaneous uploads, bulk exports and all transcription modes. An individual Unlimited account cannot be shared with other people; teams should review the separate team arrangement rather than treat one subscription as a shared login. For an interview or meeting, select the original audio language and enable speaker recognition when speaker attribution matters. Test a short, representative recording before moving a large archive. If the audio is difficult, compare a normal transcript with the Restore Audio option: the upload interface recommends restoration as a last resort for poor recordings. Listen again wherever meaning is uncertain, particularly technical terms, dates and figures. A transcription draft should not be treated as a verified medical, legal or financial record without the appropriate review. For subtitles, export SRT or VTT and inspect timing, line breaks, speaker changes and translated wording in the intended video player. For research or written work, DOCX, PDF, TXT and CSV exports support different editing and analysis workflows. Keep the original recording and record any corrections so an editor can trace a quotation back to its source. Only upload recordings you have permission to process and share the resulting text with the appropriate audience. TurboScribe is operated by Leif Erikson Ventures, LLC. Its security FAQ says media and transcripts are encrypted at rest, transcription runs on machines it owns or controls, and those files are not used to train AI models. The privacy policy places primary infrastructure in the United States and describes cloud and analytics subprocessors. It says deleted content becomes inaccessible immediately and is fully purged, including backups, within 90 days. Review that policy against your organization's data requirements before uploading confidential material. Choose TurboScribe when the main job is turning existing recordings into text or captions at predictable subscription pricing. Compare Descript when you also need to edit audio or video through the transcript, or Riverside when recording remote participants is part of the workflow. Compare the same representative file and the work needed to correct and deliver it, rather than relying on a headline accuracy percentage. |
| Category | Developer Tools | Speech-To-Text |
| Rating | No reviews | No reviews |
| Pricing | Open Source | Freemium |
| Starting Price | N/A | Free |
| Plans | — |
|
| Use Cases |
|
|
| Tags | AI gatewayPython SDKLLM routingMCP gatewayvirtual keys | audio transcriptionvideo transcriptionspeech to textspeaker recognitioncaptions |
| Features | ||
| Python SDK for direct application integration | ||
| Shared AI proxy gateway and administration UI | ||
| More than 100 documented model-provider integrations | ||
| Virtual keys, users, teams, budgets and rate limits | ||
| Spend tracking and observability integrations | ||
| Router retries, fallbacks and load balancing | ||
| MCP gateway for Streamable HTTP, SSE and stdio upstreams | ||
| Key, team and organization MCP permissions | ||
| Separate enterprise identity, audit and support features | ||
| Audio and video file transcription powered by Whisper | ||
| Three free transcripts per day, up to 30 minutes each | ||
| Unlimited files up to 10 hours or 5 GB | ||
| Speaker recognition and optional audio restoration | ||
| Transcription in 98+ languages; translation to 134+ languages | ||
| DOCX, PDF, TXT, CSV, SRT and VTT exports | ||
| Up to 50 simultaneous uploads and bulk exports on Unlimited | ||
| View BerriAI/litellm - GitHub | View TurboScribe | |
Modify This Comparison
Also Compare
Explore more head-to-head comparisons with BerriAI/litellm - GitHub and TurboScribe.