GPT-3.5 Turbo Instruct: benchmark results

Provider: OpenAI. Released 2022-11-30. Access: API.

Aggregate Rating 1520 ± 39, rank #1993 of 10552 rated models, from 16 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AI Chess Leaderboard (Continuation)1447Elo92.4
Chess Bench LLM1643Lichess-Classical-calibrated Glicko-2 rating (native signed 85.4
NCKI kspine_v368.24NCKI kspine_v3 index38.9
AI Chess Leaderboard (Reasoning)573Elo26.3
NCRI 15.249.65NCRI 15.2 index24.6
AgentDrive - Physics27.5Accuracy (%)22
AgentDrive - Scenario82.5Accuracy (%)12
METR Benchmark0.0150% Time Horizon (hours)8
AgentDrive - Comparative52.5Accuracy (%)7
AgentDrive - Hybrid5Accuracy (%)7
AgentDrive34.5Accuracy (%)6
TasteVal - Performance Multiplier0.11Performance multiplier against the expert baseline: test sco4.3

Interactive version: theaggregate.ai/model?slug=gpt-3-5-turbo-instruct · How It Works · Data refreshed daily, snapshot 2026-11-16.