opencode-model-router 1.0.3 → 1.0.4
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +11 -1
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -16,6 +16,16 @@ The agent automatically delegates via the Task tool when it recognizes the task
|
|
|
16
16
|
|
|
17
17
|
This applies both to plan-driven execution and direct ad-hoc requests. For every new user message, the orchestrator performs an intent gate, splits multi-task requests into atomic units, and routes each unit to `@fast`, `@medium`, or `@heavy`.
|
|
18
18
|
|
|
19
|
+
### Token overhead disclaimer
|
|
20
|
+
|
|
21
|
+
The injected protocol is compact, but it still adds tokens on every iteration.
|
|
22
|
+
|
|
23
|
+
- Estimated average injection: ~208 tokens per iteration
|
|
24
|
+
- Preset breakdown (default `tiers.json`): `anthropic` ~209, `openai` ~206
|
|
25
|
+
- Estimation method: `prompt_characters / 4` (rough heuristic)
|
|
26
|
+
|
|
27
|
+
Real token usage varies by tokenizer/model and any custom changes you make to `tiers.json`.
|
|
28
|
+
|
|
19
29
|
## Installation
|
|
20
30
|
|
|
21
31
|
### Option A: npm package (recommended)
|
|
@@ -25,7 +35,7 @@ Add the plugin package in your `opencode.json`:
|
|
|
25
35
|
```json
|
|
26
36
|
{
|
|
27
37
|
"plugin": [
|
|
28
|
-
"opencode-model-router@
|
|
38
|
+
"opencode-model-router@latest"
|
|
29
39
|
]
|
|
30
40
|
}
|
|
31
41
|
```
|
package/package.json
CHANGED