Provider & Model Catalogue
Every registered provider ID, its presets and models, and the ledger key that prices match against.
This page is generated from the provider registry, so it always lists the IDs
LocalRPG actually provides. Use it when writing a providers.<id> block, assigning
an agent role, or adding a providers.prices entry.
The Ledger key column is the exact provider value a metered call records
and so the value a price must use. It's also the value shown in the
Provider column of the Usage tab. A price with no model matches every model of
that key.
A ledger key is the identifier the usage ledger files a call under: the adapter that served it, and the endpoint it used. The config key you chose for a provider is your own name for it.
[!NOTE] A provider ID such as
llm:openaichatis the built-in adapter name, not the key you choose forproviders.<your-id>. The ledger key is fixed by the adapter; the config key underproviders:is yours to name.
Ledger key rules
| Family | Ledger key | Example |
|---|---|---|
| LLM | the adapter key | llm:openaichat, llm:gemini |
| Speech (TTS) | tts:gemini, tts:<name>, tts:piper@<command>, tts:http@<host> |
tts:http@localhost:8880 |
| Transcription (STT) | stt:whisper-http@<host>, stt:whisper-cli@<command> |
stt:whisper-http@localhost:8000 |
| Image | image:gemini, image:http@<host>, image:cli@<command> |
image:http@127.0.0.1:8188 |
| Embedding | embedding:openai, embedding:gemini, embedding:builtin |
embedding:gemini@default |
LLM providers
| Provider ID | Tier | Source | Ledger key | Presets and models |
|---|---|---|---|---|
llm:cli |
local-server | cli | llm:cli@llama-cli |
claude-cli, llama-cli |
llm:gemini |
cloud | gemini | llm:gemini |
gemini-3.8-flash |
llm:inworld |
cloud | http | llm:inworld |
inworld/compare-frontier-models |
llm:narrative-oracle |
offline-basic | builtin | llm:narrative-oracle |
narrative-oracle |
llm:openaichat |
cloud | http | llm:openaichat@localhost:11434 |
default, gpt-4, llama3.2 |
Speech (TTS) providers
| Provider ID | Tier | Source | Ledger key | Presets and models |
|---|---|---|---|---|
tts:cartesia |
cloud | http | tts:cartesia |
sonic-3.6 |
tts:elevenlabs |
cloud | http | tts:elevenlabs |
eleven_multilingual_v2 |
tts:fish-audio |
local-server | http | tts:http@localhost:8091 |
fishaudio/s2-pro |
tts:gemini |
cloud | gemini | tts:gemini |
gemini-2.5-flash-preview-tts, gemini-2.5-pro-preview-tts, gemini-3.1-flash-tts-preview, gemini-3.8-flash-lite-tts, gemini-3.8-flash-tts |
tts:http |
local-server | http | tts:http@localhost:8880 |
alltalk, kokoro, tts-1 |
tts:inworld |
cloud | http | tts:inworld |
inworld-tts-2 |
tts:native-os |
offline-basic | builtin | tts:native-os |
native-os |
tts:piper |
local-server | cli | tts:piper@piper |
piper |
tts:sherpa-onnx |
offline-neural | builtin | tts:sherpa-onnx |
sherpa-onnx |
Transcription (STT) providers
| Provider ID | Tier | Source | Ledger key | Presets and models |
|---|---|---|---|---|
stt:cartesia |
cloud | http | stt:cartesia |
ink-whisper |
stt:inworld |
cloud | http | stt:inworld |
inworld/inworld-stt-1 |
stt:web-speech |
offline-basic | builtin | not reported |
web-speech |
stt:whisper-cli |
local-server | cli | stt:whisper-cli@whisper-cli |
whisper-cli |
stt:whisper-http |
local-server | http | stt:whisper-http@localhost:8000 |
whisper-1 |
Image providers
| Provider ID | Tier | Source | Ledger key | Presets and models |
|---|---|---|---|---|
image:cli |
local-server | cli | image:cli@sd |
sd-cli |
image:gemini |
cloud | gemini | image:gemini |
gemini-2.5-flash-image, gemini-3-pro-image, gemini-3.1-flash-image, gemini-3.1-flash-lite-image, imagen-3.0-fast-generate-001, imagen-3.0-generate-002 |
image:http |
local-server | http | image:http@127.0.0.1:8188 |
automatic1111, comfyui, dall-e-3, stablediffusion |
image:procedural-art |
offline-basic | builtin | image:procedural-art |
procedural-art |
Embedding providers
| Provider ID | Tier | Source | Ledger key | Presets and models |
|---|---|---|---|---|
embedding:gemini |
cloud | gemini | embedding:gemini |
text-embedding-004 |
embedding:onnx |
offline-neural | builtin | embedding:onnx |
onnx |
embedding:openai |
cloud | http | embedding:openai |
nomic-embed-text, text-embedding-3-small |
Built-in default prices
These rates apply when no providers.prices entry matches, and only when the
ledger key matches exactly. A row with no model applies to every model of that
key. Add a config entry to override or extend them. Local and built-in adapters
are absent on purpose: they aren't metered.
| Ledger key | Model | Input (per 1M) | Output (per 1M) | Per character | Per request |
|---|---|---|---|---|---|
llm:gemini |
gemini-3.8-flash |
750000 | 3750000 | 0 | 0 |
llm:gemini |
gemini-3.5-flash-lite |
300000 | 2500000 | 0 | 0 |
llm:gemini |
gemini-3.1-flash-lite |
250000 | 1500000 | 0 | 0 |
llm:gemini |
gemini-2.5-flash |
300000 | 2500000 | 0 | 0 |
llm:gemini |
gemini-2.5-flash-lite |
100000 | 400000 | 0 | 0 |
llm:gemini |
gemini-2.0-flash |
150000 | 600000 | 0 | 0 |
llm:gemini |
any |
750000 | 3750000 | 0 | 0 |
llm:openaichat@api.openai.com |
gpt-4o |
2500000 | 10000000 | 0 | 0 |
llm:openaichat@api.openai.com |
gpt-4o-mini |
150000 | 600000 | 0 | 0 |
llm:openaichat@api.openai.com |
any |
150000 | 600000 | 0 | 0 |
tts:elevenlabs |
any |
0 | 0 | 80 | 0 |
tts:http@api.openai.com |
any |
0 | 0 | 15 | 0 |
image:gemini |
imagen-3.0-generate-002 |
0 | 0 | 0 | 40000 |
image:gemini |
imagen-3.0-fast-generate-001 |
0 | 0 | 0 | 20000 |
embedding:openai@api.openai.com |
text-embedding-3-small |
20000 | 0 | 0 | 0 |
embedding:openai@api.openai.com |
text-embedding-3-large |
130000 | 0 | 0 | 0 |
Prices are micros: one millionth of a currency unit, so 2500000 is 2.50.