v1.9.0
v1.9.0 - Esperanto 2.22 & New Audio Providers
Added 10
- Mistral Voxtral provider for speech-to-text and text-to-speech
- Deepgram provider for text-to-speech with Aura voices
- xAI provider for text-to-speech
- Google speech-to-text and text-to-speech providers
- Vertex text-to-speech provider
- ElevenLabs speech-to-text provider (Scribe)
- Per-credential Ollama num_ctx override to configure context window in Settings → API Keys
- OPEN_NOTEBOOK_EMBEDDING_BATCH_SIZE environment variable to tune embedding batch size
- CORS_ORIGINS environment variable to configure allowed API origins
- OPEN_NOTEBOOK_MIN_CHUNK_SIZE environment variable to filter degenerate tiny chunks
Changed 3
- Speech-to-text connection tests now transcribe a real speech clip instead of silence
- Ollama default context window (num_ctx) is now 8192 instead of 128000
- Google embedding default model is now gemini-embedding-001
Fixed 7
- OpenRouter LLM and embedding requests now send proper JSON body
- Clear errors when OpenAI-compatible endpoints return null embeddings instead of opaque TypeError
- Streaming tool calls return proper objects across Anthropic, Google, Vertex, and Ollama
- base_url trailing slashes normalized to fix double-slash 301 errors for Ollama and self-hosted
- Ollama thinking models merge reasoning content correctly
- Model discovery honors custom base_url for LiteLLM, vLLM, and proxies
- URL-source embedding crash on single-character HTML fragments
✨ Highlights
- Upgraded to Esperanto 2.22.0, unlocking new providers and inheriting several upstream fixes.
- New audio providers:
- Mistral Voxtral — speech-to-text and text-to-speech (reuses your Mistral credential)
- Deepgram — text-to-speech (Aura voices) — new provider
- xAI — text-to-speech
- Completed the matrix: Google STT/TTS, Vertex TTS, ElevenLabs STT (Scribe)
- Per-credential Ollama
num_ctxoverride — configure the context window for Ollama models in Settings → API Keys (translated across all 13 locales). - Better speech-to-text connection tests — they now transcribe a real speech clip instead of silence, so a passing test shows actual text.
⚠️ Upgrade Notes (behavior changes inherited from Esperanto)
- Ollama default context window (
num_ctx) is now 8192 (was 128000) to avoid out-of-memory errors on consumer GPUs. If you ingest large documents with Ollama, set a highernum_ctxon the Ollama credential. - Google embedding default model is now
gemini-embedding-001— the previous default,text-embedding-004, was removed from Google's API. If you used Google embeddings with the old default, re-create the model and re-embed your content (the embedding dimensions changed).
🛠️ Inherited Esperanto fixes
- OpenRouter LLM & embedding requests now send a proper JSON body.
- Clear errors (instead of an opaque
TypeError) when OpenAI-compatible endpoints return null embeddings. - Streaming tool calls return proper objects across Anthropic, Google, Vertex, and Ollama.
base_urltrailing slashes normalized (fixes double-slash 301s for Ollama / self-hosted).- Ollama "thinking" models (e.g. Qwen) merge reasoning content correctly.
- Model discovery honors a custom
base_url(LiteLLM/vLLM/proxies).
➕ Also in this release
OPEN_NOTEBOOK_EMBEDDING_BATCH_SIZEenv var to tune embedding batch size (#735)CORS_ORIGINSenv var to configure allowed API origins (#585, #597, #730)OPEN_NOTEBOOK_MIN_CHUNK_SIZEenv var to filter degenerate tiny chunks- Fixed URL-source embedding crash on single-character HTML fragments (#764)
- README Provider Support Matrix synced with actual capabilities
Full changelog: see CHANGELOG.md · compare v1.8.5...v1.9.0