TruthfulQA
AI benchmark results and model performance
Type: percentageFinance
Recorded results
24
Average score
63.20%
Score range
50.30% - 77.50%
Leading model
77.50% - Phi 3.5 MoE instruct
Scores over time
Individual benchmark scores plotted by date.
Model results
| Organisation | Models | Reported | Top score | Info | Self reported | Source |
|---|---|---|---|---|---|---|
| Phi 3.5 MoE instruct | 23 Aug 2024 | 77.50% | - | Yes | Source | |
| Granite 3.0 8B Instruct | - | 66.86% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14) | Yes | Source | |
| Granite 3.1 8B Instruct | - | 66.86% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14) | Yes | Source | |
| Granite 3.2 8B Instruct | - | 66.86% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14) | Yes | Source | |
| Granite 3.2 8B Instruct Preview | - | 66.86% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4687; benches=14) | Yes | Source | |
| Granite 3.3 2B Instruct | 16 Apr 2025 | 66.86% | inferred family alias from granite-3.3-8b-instruct (score=0.3627; benches=14) | Yes | Source | |
| Granite 3.3 8B Instruct | 16 Apr 2025 | 66.86% | - | Yes | Source | |
| Granite Guardian 3.0 8B | - | 66.86% | inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14) | Yes | Source | |
| Granite Guardian 3.1 8B | - | 66.86% | inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14) | Yes | Source | |
| Granite Guardian 3.3 8B | - | 66.86% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.5071; benches=14) | Yes | Source | |
| Granite Speech 3.2 8B | - | 66.86% | inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14) | Yes | Source | |
| Granite Speech 3.3 8B | - | 66.86% | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.5071; benches=14) | Yes | Source | |
| Phi 4 Mini | 01 Feb 2025 | 66.40% | - | Yes | Source | |
| Phi 3 Mini 128K Instruct | - | 64% | inferred family alias from phi-3.5-mini-instruct (score=0.3533; benches=31) | Yes | Source | |
| Phi 3.5 mini instruct | 23 Aug 2024 | 64% | - | Yes | Source | |
| Llama 3.1 Nemotron 70B Instruct | 01 Oct 2024 | 58.63% | - | Yes | Source | |
| Qwen 14B | - | 58.40% | inferred family alias from qwen-2.5-14b-instruct (score=0.3060; benches=16) | Yes | Source | |
| Granite 4.0 Micro | 02 Oct 2025 | 58.10% | inferred high-confidence family alias from granite-4.0-tiny-preview (score=0.4700; benches=12) | Yes | Source | |
| Granite 4.0 Small | 02 Oct 2025 | 58.10% | inferred high-confidence family alias from granite-4.0-tiny-preview (score=0.4700; benches=12) | Yes | Source | |
| Granite 4.0 Tiny | 02 Oct 2025 | 58.10% | inferred alias from granite-4.0-tiny-preview | Yes | Source | |
| Granite 4.0 Tiny Preview | 02 May 2025 | 58.10% | - | Yes | Source | |
| Qwen 2 Math 72B | - | 54.80% | inferred high-confidence family alias from qwen2-72b-instruct (score=0.4667; benches=17) | Yes | Source | |
| Qwen 2 Math RM 72B | - | 54.80% | inferred family alias from qwen2-72b-instruct (score=0.3917; benches=17) | Yes | Source | |
| Mistral Nemo 12B | 18 Jul 2024 | 50.30% | inferred family alias from mistral-nemo-instruct-2407 (score=0.3250; benches=8) | Yes | Source |
