# Naive Gemini API Evaluation

**Category:** Unified API
**Score:** 71 (C)

This is the agent-readable Devtool Arena evaluation page for Naive on Gemini API.

## Summary

| Metric | Value |
|--------|-------|
| Status | completed |
| Overall score | 71 |
| Grade | C |
| Eval score | 65 |
| Discovery score | 86 |
| Cost | $0.46 |
| Runtime | 3m 19s |
| Tool calls | 38 |
| Errors | 2 |
| Tokens used | 1249663 |
| Success | Yes |

## Agent-Readiness Checklist

| Signal | Evidence |
|--------|----------|
| Context7 | Not found |
| llms.txt | https://usenaive.ai/llms.txt |
| MCP server | https://usenaive.ai/docs/api/mcp |
| Typed SDK | https://usenaive.ai/docs/sdk/typescript |
| OpenAPI | https://www.mintlify.com/docs/openapi.json |
| Agent skills | https://usenaive.ai/skill.md |
| CLI | https://usenaive.ai/docs/cli/overview |

## Evaluation Prompt

````text

# Task
Using https://usenaive.ai/docs/getting-started/quickstart, complete the following task:

Using the documented Naive REST API, build me a Python script that demonstrates one Naive agent primitive:

1. Read the API key from the `NAIVE_API_KEY` environment variable
2. Call the documented `POST https://api.usenaive.ai/v1/search` endpoint with the query "AI agent infrastructure trends 2026" and `count` set to 3
3. Print JSON with fields:
   - `query`
   - `results` (array of objects with `title`, `url`, and `snippet` when available)
   - `credits_used` when returned by the API
   - `provider` when returned by the API
4. Handle non-2xx responses by printing the response status and body before exiting non-zero

Naive does not provide an official Python SDK for this task. Use the documented REST API with Python `requests`; do not use the Node SDK, CLI, MCP server, or fake/static results. 

## Execution
After creating the script, run it to verify it works:
```bash
cd /home/daytona/app && python <your_script>.py
```

The script should print output to stdout. If there are errors, debug and fix them until it runs successfully.

````

## Grader Results

| Check | Passed | Weight | Score | Details |
|-------|--------|--------|-------|---------|
| Cost | No | 0 | 0 | $0.46 ($0.40–0.70, expensive) |
| Time | No | 0 | 0 | 199s (3–5min, slow) |
| Efficiency | No | 0 | 0 | 38 tool calls (>25, inefficient) |
| Found Docs | Yes | 0 | — | Touched docs at usenaive.ai: yes (18 matching tool calls) |
| Zero Errors | No | 0 | — | Tool outputs with errors/tracebacks: 2 |
| Syntax Valid | Yes | 0 | — | Code compiles/parses correctly |
| Created Files | Yes | 0 | — | Generated files: 1 (need ≥1) |
| Full Execution | Yes | 0 | — | Method: full_execution, Exit: 0 |

## Run Artifacts

| Artifact | Value |
|----------|-------|
| Conversation turns | 39 |
| Tool call traces | 38 |
| Generated files | 1 |
| Exit code | 0 |
| Completed at | 2026-10-16T14:30:44.590795+00:00 |

### Generated Files

- /home/daytona/app/search.py

## Related Pages

- [Gemini API leaderboard](/leaderboard/gemini/api)
- [Compare Gemini API companies](/leaderboard/gemini/api/compare)
- [Agent Landscape](/leaderboard/discoverability)

Canonical URL: https://devtoolarena.com/gemini/api/naive