4071551476
- real-fit-engine.py: refactored to support --from-report, improved Ollama v1/chat/completions compatibility, agent name normalization - run-focused-eval.py: run evaluations for specific agent/model pairs from CLI - test_ollama_minimal.py/test_real_api.py: Ollama API connectivity tests - real-fit-architecture.md: architecture overview document - tests/scripts/: E2E landing test, analytics capture, evolution heatmap verification - Remove real-fit-recalc.py (superseded by --from-report flag)
60 lines
2.0 KiB
Python
60 lines
2.0 KiB
Python
#!/usr/bin/env python3
|
|
import urllib.request, json, os, time
|
|
|
|
def call_ollama_real(model_short, system_prompt, user_prompt):
|
|
key = os.environ.get("OLLAMA_KEY", "")
|
|
host = "https://ollama.com/v1"
|
|
|
|
payload = json.dumps({
|
|
"model": model_short,
|
|
"messages": [
|
|
{"role": "system", "content": system_prompt},
|
|
{"role": "user", "content": user_prompt}
|
|
],
|
|
"temperature": 0.3,
|
|
"max_tokens": 2048
|
|
}).encode()
|
|
|
|
req = urllib.request.Request(
|
|
f"{host}/chat/completions",
|
|
data=payload,
|
|
headers={
|
|
"Content-Type": "application/json",
|
|
"Authorization": f"Bearer {key}" if key else "Bearer",
|
|
"User-Agent": "Mozilla/5.0"
|
|
},
|
|
method="POST"
|
|
)
|
|
|
|
start = time.time()
|
|
try:
|
|
with urllib.request.urlopen(req, timeout=120) as resp:
|
|
data = json.loads(resp.read().decode())
|
|
text = data.get("choices", [{}])[0].get("message", {}).get("content", "")
|
|
usage = data.get("usage", {})
|
|
elapsed = int((time.time() - start) * 1000)
|
|
print(f"Status: {resp.status}")
|
|
print(f"Latency: {elapsed}ms")
|
|
print(f"Tokens: prompt={usage.get('prompt_tokens')}, completion={usage.get('completion_tokens')}")
|
|
return text
|
|
except urllib.error.HTTPError as e:
|
|
body = e.read().decode()[:200]
|
|
print(f"HTTP Error: {e.code} {e.reason}")
|
|
print(f"Body: {body}")
|
|
return ""
|
|
except Exception as e:
|
|
print(f"Error: {e}")
|
|
return ""
|
|
|
|
if __name__ == "__main__":
|
|
print("=== Test real Ollama API ===")
|
|
text = call_ollama_real(
|
|
"kimi-k2.6",
|
|
"You are a code reviewer. Find bugs.",
|
|
"Review: def f(x): return x+1"
|
|
)
|
|
print(f"\nResponse (first 300 chars):\n{text[:300]}")
|
|
print(f"\nTotal length: {len(text)} chars")
|
|
print(f"Keyword 'naming' in response: {'naming' in text.lower()}")
|
|
print(f"Keyword 'return' in response: {'return' in text.lower()}")
|