Model results

IFBench

Choose a shared evaluation protocol to compare model scores. Versions, sources and configurations stay separate.

Atlas snapshot captured 10/19/2026, 10:20:28 AM UTC. This is an import timestamp, not the evaluation date. Coverage may be incomplete.

Official source

← All evaluation groups

IFBench

Source: Artificial Analysis
Version: Not reported
Measurement date: Not reported
Metric: % · Higher is better
Evaluation scope: Instruction-following evaluation; observed 2026-10-19; result date not supplied

Exact score ties share a position. Positions do not establish statistical significance. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.

#1MiniMax M3
minimax/minimax-m3-20260531
82.9%
View modelEvidence independent
#2Nemotron 3 Ultra
nvidia/nemotron-3-ultra-550b-a55b-20260604
81.4%
View modelEvidence independent
#3MiMo-V2.5-Pro
xiaomi/mimo-v2.5-pro-20260422
79.9%
View modelEvidence independent
#4Command A
cohere/command-a-03-2025
73.9%
View modelEvidence independent
#4Command A+
cohere/command-a-plus-05-2026
73.9%
View modelEvidence independent
#6GPT-5.6 Sol
openai/gpt-5.6-sol-20260709
72.7%
View modelEvidence independent
#7GPT-5.6 Terra
openai/gpt-5.6-terra-20260709
71.2%
View modelEvidence independent
#8gpt-oss-120b
openai/gpt-oss-120b
69%
View modelEvidence independent
#9Mistral Medium 3
mistralai/mistral-medium-3
68.8%
View modelEvidence independent
#9Mistral Medium 3.1
mistralai/mistral-medium-3.1
68.8%
View modelEvidence independent
#9Mistral Medium 3.5
mistralai/mistral-medium-3.5-20260430
68.8%
View modelEvidence independent
#12MiMo-V2.5
xiaomi/mimo-v2.5-20260422
67.1%
View modelEvidence independent
#13Claude Fable 5
anthropic/claude-fable-5
63.5%
View modelEvidence independent
#14MiMo-V2-Flash
xiaomi/mimo-v2-flash
39.9%
View modelEvidence independent
#15Phi 4
microsoft/phi-4
23.5%
View modelEvidence independent