Back to leaderboard

Compare companies

Side-by-side API results for companies evaluated by Codex. Compare score, rank, discoverability, cost, errors, runtime, and the underlying eval artifacts.

Model
Claude CodeClaude CodeCodexCodex
Tooling
APICLIMCP

Choose 2-4 companies

Selected
Deepgram
OpenRouter
Showing the first two companies on mobile. Use a wider screen to compare 3 companies.
Mobile compare is limited to two companies. Use a wider screen to compare all 3.
Deepgram

Deepgram

Voice STT

OpenRouter

OpenRouter

Inference

Summary

Compare status

Comparable
Comparable

Overall score

86
Best
79

Grade

B
B

Category

Voice STT
Inference

Overall rank

#3
Best
#20

Category rank

#1
Best
#2

Full eval

Open report
Open report
Agent Performance

Eval score

80%
Best
76%

Task success

Yes
Yes

Cost

$0.2314
$0.1847
Best

Tool calls

16
Best
25

Errors

1
Best
4

Runtime

46s
Best
1m 7s

Tokens used

307,782
Best
607,456

Exit code

0
Best
0
Best
Discoverability

Discovery score

100%
Best
86%

Context7

Yes
Yes

llms.txt

Yes
Yes

MCP Server

Yes
Yes

Typed SDK

Yes
Yes

OpenAPI

Yes
Yes

Agent Skills

Yes
Yes

CLI

Yes
Best
No
Eval Details

Integration task

# Task Using https://developers.deepgram.com/docs, complete the following task: Using the Deepgram Python SDK, build me a script that transcribes a sample audio file: 1. Install...
# Task Using https://openrouter.ai/docs, complete the following task: Using OpenRouter, build me a script that extracts structured data from this invoice and returns JSON: --- I...

Generated files

1
1

Trace tool calls

16
25

Completed

9/27/2026
9/27/2026

Stderr

None captured
None captured