Phi-3.5 on TruthfulQA: score, rank and what it means
Phi-3.5 scores 74.8 on TruthfulQA, placing it #336 of 600 models in the index. The benchmark average is 73.6 and the leader is GPT-5.1 Max at 91.1.
Score distribution
The chart shows the top 10 models on TruthfulQA, with Phi-3.5 highlighted. Longer bars mean higher scores.
Top 10 leaderboard
| # | Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|---|
| 1 | GPT-5.1 Max | OpenAI | 2000000 | $20.50 | 87.9 |
| 2 | Gemini 3 Pro Reasoning | Google DeepMind | 200000 | $19.69 | 87.5 |
| 3 | GPT-5.1 Reasoning | OpenAI | 400000 | $20.27 | 88.3 |
| 4 | GPT-5.2 Pro | OpenAI | 2000000 | $21.32 | 89.6 |
| 5 | GPT-5.2 | OpenAI | 2000000 | $19.71 | 88.0 |
| 6 | Gemini 3 Pro Max | Google DeepMind | 2000000 | $20.57 | 87.9 |
| 7 | Grok 4 Reasoning | xAI | 2000000 | $19.67 | 87.8 |
| 8 | Qwen3-Max Max | Alibaba Cloud | 400000 | $18.53 | 85.9 |
| 9 | GPT-5.1 | OpenAI | 2000000 | $19.01 | 86.0 |
| 10 | GPT-5.1 Turbo | OpenAI | 2000000 | $19.19 | 86.1 |
What this score means
On TruthfulQA, Phi-3.5 sits at #336 of 600. The leader, GPT-5.1 Max, scores 91.1, so the gap to the top is 16.3 points. The median model in this index scores 76.3, which puts Phi-3.5 below the midpoint.
Microsoft models average 67.5 on this benchmark. Phi-3.5 is 7.3 points above that provider line, so it is a stronger result than the average Microsoft model.
Among open-weight models, the average is 72.3. Phi-3.5 is 2.5 points above that group. This is the more useful comparison if you are shortlisting within the same tier rather than across the whole index.
Read a 74.8 as a relative signal, not an absolute guarantee. Benchmark scores depend on prompt format, evaluation harness and contamination controls. Use this page to narrow the shortlist, then test the model on your own data.
Price context
At $2.81 per million input tokens, Phi-3.5 is below the median input price of $4.00 across the 600 models scored on TruthfulQA. The models immediately around it in the ranking are shown below so you can compare score and price together.
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Mixtral Groq Base | Groq | 32768 | $2.66 | 68.4 |
| Firefunction Pro | Fireworks AI | 16384 | $9.05 | 72.9 |
| Grok 4 Nano | xAI | 16384 | $8.97 | 72.7 |
| Hunyuan-Large Base | Tencent | 200000 | $3.64 | 74.0 |
| Phi-3.5 | Microsoft | 16384 | $2.81 | 70.9 |
| Auto Reasoning | OpenRouter | 65536 | $8.39 | 71.5 |
| Llama 3 Baseten Base | Baseten | 16384 | $3.16 | 72.4 |
Other benchmarks for Phi-3.5
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 76.1 | 304 | 49th | |
| GPQA Diamond | 67.0 | 331 | 45th | |
| SWE-bench Verified | 60.9 | 308 | 49th | |
| HumanEval | 76.2 | 327 | 46th | |
| MATH-500 | 75.4 | 316 | 47th | |
| AIME 2025 | 60.6 | 383 | 36th | |
| LiveCodeBench | 59.8 | 371 | 38th | |
| IFEval | 77.1 | 377 | 37th | |
| HellaSwag | 82.9 | 323 | 46th | |
| MMMU | 68.8 | 328 | 45th | |
| Arena Elo | 1356.7 | 350 | 42th |
FAQ
What score does Phi-3.5 get on TruthfulQA?
Phi-3.5 scores 74.8 on TruthfulQA in this index. The benchmark average is 73.6 and the top score is 91.1.
How does Phi-3.5 rank on TruthfulQA?
It ranks #336 of 600 models, in the 44th percentile. The leader is GPT-5.1 Max at 91.1.
Is Phi-3.5 good at TruthfulQA?
It is below the median for this benchmark, so test it carefully before relying on it for this capability. The result should be read alongside its price and context window.
How is the TruthfulQA score calculated?
TruthfulQA is a safety benchmark. Scores in this index are modelled estimates on a 0–100 scale (or Elo for Arena), derived from capability tier and benchmark difficulty. They are not official leaderboard results.
What does Phi-3.5 cost?
Phi-3.5 costs $2.81 per million input tokens and $14.99 per million output tokens.