Which model fomo2 picks for each kind of LLM call. Override per task
via llm.task_routing in config/default.yaml,
or per-feed via llm_task.
| Task | Provider | Model | Overridden |
|---|---|---|---|
dedup |
ollama | gpt-oss:20b-cloud | no |
default |
ollama | gpt-oss:120b-cloud | no |
embedding |
gemini | gemini-embedding-001 | no |
enrich |
ollama | gpt-oss:20b-cloud | no |
goal_default |
ollama | gpt-oss:120b-cloud | no |
goal_pro |
ollama | kimi-k2.6 | yes |
narrative |
ollama | gpt-oss:20b-cloud | no |
prediction_scoring |
ollama | qwen3-coder:480b-cloud | no |
super_report |
ollama | kimi-k2.6 | no |
validation |
ollama | gpt-oss:120b-cloud | no |
watcher |
ollama | qwen3-coder:480b-cloud | no |
wizard |
ollama | gpt-oss:120b-cloud | no |
See Smart LLM Picker for OpenRouter rankings and recommended Ollama matches.
Exchange rate: 1 USD = 0.86 EUR
| Model | Input (USD) | Output (USD) | Embedding (USD) |
|---|---|---|---|
| claude-haiku-4-5-20251001 | $0.800 | $4.000 | n/a |
| claude-sonnet-4-20250514 | $3.000 | $15.000 | n/a |
| deepseek-v3.1:671b-cloud | free | free | n/a |
| embeddinggemma:300m | free | free | n/a |
| gemini-2.0-flash | $0.100 | $0.400 | n/a |
| gemini-2.0-flash-lite | $0.075 | $0.300 | n/a |
| gemini-2.5-flash | $0.150 | $0.600 | n/a |
| gemini-2.5-pro | $1.250 | $10.000 | n/a |
| gemini-embedding-001 | free | free | $0.150 |
| google/gemini-2.0-flash-exp:free | free | free | n/a |
| gpt-4o | $2.500 | $10.000 | n/a |
| gpt-4o-mini | $0.150 | $0.600 | n/a |
| gpt-oss:120b-cloud | free | free | n/a |
| gpt-oss:20b-cloud | free | free | n/a |
| kimi-k2.6 | free | free | n/a |
| qwen3-coder:480b-cloud | free | free | n/a |
| text-embedding-004 | free | free | $0.100 |
| text-embedding-3-small | free | free | $0.020 |