SpeechMap Compliance: leaderboard

Metric: % Requests Completed. Source: speechmap.ai. 382 models tracked.

Top models

#ModelScore
1Grok 498.3
2Mistral Large 398.2
3Grok 4.20 (Reasoning)98.2
4Grok 4.1 Fast (Non-reasoning)97.9
5Grok 396.1
6Grok 2 (1212)95.9
7Grok 4.1 Fast (Reasoning)95.7
8GPT-4 (0314)95.2
9GPT-3.5 Turbo (0125)94.5
10Mistral Small 3.293.7
11Grok Beta93.7
12Ministral 3 14B93.3
13Grok 4.593.1
14Magistral Medium92.6
15Grok 3 Mini Beta92.5

Interactive version: theaggregate.ai/benchmark?slug=speechmap-compliance · How It Works · Data refreshed daily, snapshot 2026-09-05.