max_tokens per vaihe: manageri 200, koodari 512, testaaja 200, QA 512, DevOps 256
Hub ja natiivisolmu tukevat nyt max_tokens-kenttää API-pyynnöissä. Pipeline-vaiheet käyttävät sopivan kokoisia token-rajoja. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
@@ -947,6 +947,8 @@ struct ChatCompletionRequest {
|
||||
model: String,
|
||||
prompt: String,
|
||||
task_id: String,
|
||||
#[serde(default)]
|
||||
max_tokens: Option<u64>,
|
||||
}
|
||||
|
||||
#[derive(serde::Serialize)]
|
||||
@@ -1069,12 +1071,15 @@ async fn api_chat_completions(
|
||||
state.node_busy.lock().unwrap().insert(target_node_id);
|
||||
state.pending_task_ids.lock().unwrap().insert(payload.task_id.clone());
|
||||
|
||||
let msg = serde_json::json!({
|
||||
let mut msg = serde_json::json!({
|
||||
"type": "llm_prompt",
|
||||
"prompt": payload.prompt,
|
||||
"model": payload.model,
|
||||
"task_id": payload.task_id,
|
||||
});
|
||||
if let Some(mt) = payload.max_tokens {
|
||||
msg.as_object_mut().unwrap().insert("max_tokens".to_string(), serde_json::json!(mt));
|
||||
}
|
||||
|
||||
// Odotuskanava valmiiksi (solmu palauttaa tuloksen stats_tx kautta)
|
||||
let mut rx = state.stats_tx.subscribe();
|
||||
|
||||
Reference in New Issue
Block a user