feat: update models to claude-opus-5/sonnet-5/haiku-4-5-20251001 (tested live on 9router)

- PRIMARY: openai/claude-opus-5 (was openai/claude-opus-4-8)
- FALLBACKS: added openai/claude-sonnet-5, openai/claude-haiku-4-5-20251001
- Verified all three return valid responses via raw HTTP to 9router
- Updated run_server.py, health-check.py, setup_all.py
This commit is contained in:
asepharyana
2026-08-20 11:30:10 +07:00
parent 54cec6bf0c
commit 017656d97b
3 changed files with 7 additions and 7 deletions
+2 -2
View File
@@ -19,8 +19,8 @@ from pathlib import Path
BWS_SECRET_ID = "2aef2194-971d-4dae-99dd-b49a0041f97c" BWS_SECRET_ID = "2aef2194-971d-4dae-99dd-b49a0041f97c"
ROUTER_BASE = "https://9router.asepharyana.my.id/v1" ROUTER_BASE = "https://9router.asepharyana.my.id/v1"
PRIMARY = "openai/claude-opus-4-8" PRIMARY = "openai/claude-opus-5"
FALLBACKS = ["openai/ATLAS", "openai/gemini", "openai/text", "openai/deepseek-v4-flash-free"] FALLBACKS = ["openai/claude-sonnet-5", "openai/claude-haiku-4-5-20251001", "openai/ATLAS", "openai/gemini", "openai/text", "openai/deepseek-v4-flash-free"]
# Caddy 9router route is now response_header_timeout 120s / read 300s. # Caddy 9router route is now response_header_timeout 120s / read 300s.
# LLM combo TTFT often 30-40s+. Give the check room to complete. # LLM combo TTFT often 30-40s+. Give the check room to complete.
HTTP_TIMEOUT = 150 HTTP_TIMEOUT = 150
+3 -3
View File
@@ -26,10 +26,10 @@ os.environ["OPENAI__API_BASE"] = os.environ.get(
"OPENAI_API_BASE", "https://omniroute.imrnes.team/v1" "OPENAI_API_BASE", "https://omniroute.imrnes.team/v1"
) )
os.environ["OPENAI__KEY"] = omni_key os.environ["OPENAI__KEY"] = omni_key
os.environ["CONFIG__MODEL"] = os.environ.get("PR_AGENT_MODEL", "openai/claude-opus-4-8") os.environ["CONFIG__MODEL"] = os.environ.get("PR_AGENT_MODEL", "openai/claude-opus-5")
os.environ["CONFIG__FALLBACK_MODELS"] = os.environ.get( os.environ["CONFIG__FALLBACK_MODELS"] = os.environ.get(
"PR_AGENT_FALLBACK_MODELS", "PR_AGENT_FALLBACK_MODELS",
'["openai/ATLAS","openai/gemini","openai/text","openai/deepseek-v4-flash-free"]', '["openai/claude-sonnet-5","openai/claude-haiku-4-5-20251001","openai/ATLAS","openai/gemini","openai/text","openai/deepseek-v4-flash-free"]',
) )
os.environ["CONFIG__CUSTOM_MODEL_MAX_TOKENS"] = os.environ.get( os.environ["CONFIG__CUSTOM_MODEL_MAX_TOKENS"] = os.environ.get(
"PR_AGENT_MAX_TOKENS", "128000" "PR_AGENT_MAX_TOKENS", "128000"
@@ -261,7 +261,7 @@ if __name__ == "__main__":
port = int(os.environ.get("PORT", "3000")) port = int(os.environ.get("PORT", "3000"))
print(f"PR-Agent GitHub App server starting...") print(f"PR-Agent GitHub App server starting...")
print(f" App ID: {os.environ.get('GITHUB_APP_ID', '')}") print(f" App ID: {os.environ.get('GITHUB_APP_ID', '')}")
print(f" Model: {os.environ.get('PR_AGENT_MODEL', 'openai/claude-opus-4-8')} via omniroute") print(f" Model: {os.environ.get('PR_AGENT_MODEL', 'openai/claude-opus-5')} via omniroute")
print(f" Endpoint: /api/v1/github_webhooks") print(f" Endpoint: /api/v1/github_webhooks")
print(f" Analytics: {ANALYTICS_DIR}") print(f" Analytics: {ANALYTICS_DIR}")
print(f" Port: {port}") print(f" Port: {port}")
+2 -2
View File
@@ -78,8 +78,8 @@ with open(BASE_DIR / ".secrets.toml", "w") as f:
# ── 4. Create PR-Agent config ── # ── 4. Create PR-Agent config ──
config_toml = """[config] config_toml = """[config]
model = "openai/claude-opus-4-8" model = "openai/claude-opus-5"
fallback_models = ["openai/ATLAS", "openai/gemini", "openai/text", "openai/deepseek-v4-free"] fallback_models = ["openai/claude-sonnet-5", "openai/claude-haiku-4-5-20251001", "openai/ATLAS", "openai/gemini", "openai/text", "openai/deepseek-v4-flash-free"]
custom_model_max_tokens = 128000 custom_model_max_tokens = 128000
git_provider = "github" git_provider = "github"
publish_output = true publish_output = true