Phi-3.5 1.5B: pricing, context window and benchmarks
Phi-3.5 1.5B is a small open-weight model from Microsoft. It offers a 32,768-token context window, costs $0.25 per million input tokens and $1.05 per million output tokens, and averages 44.5 across the index's benchmark suite.
Pricing
Phi-3.5 1.5B costs $0.25 per million input tokens and $1.05 per million output tokens. A blended 3:1 input-to-output mix works out to about $0.45 per million tokens. Cached input is listed at $0.03 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $0.25 | per 1M tokens |
| Output | $1.05 | per 1M tokens |
| Cached input | $0.03 | per 1M tokens |
| Blended 3:1 | $0.45 | estimate per 1M tokens |
| 1M in + 1M out | $1.30 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 51.3 | 587 | 2th | |
| GPQA Diamond | 38.3 | 591 | 2th | |
| SWE-bench Verified | 27.8 | 585 | 3th | |
| HumanEval | 48.9 | 582 | 3th | |
| MATH-500 | 43.5 | 585 | 3th | |
| AIME 2025 | 32.8 | 577 | 4th | |
| LiveCodeBench | 27.7 | 588 | 2th | |
| IFEval | 58.8 | 579 | 4th | |
| TruthfulQA | 53.7 | 580 | 3th | |
| HellaSwag | 65.1 | 583 | 3th | |
| MMMU | 41.9 | 589 | 2th | |
| Arena Elo | 1145.1 | 585 | 3th |
What Phi-3.5 1.5B is good for
Phi-3.5 1.5B is a small model. That makes it a good fit for classification, extraction, routing, summarization and other high-volume jobs where latency and cost dominate. It is not the right pick for open-ended frontier reasoning, but at $0.25 per million input tokens it can handle a large share of production traffic.
Strengths
- Input price of $0.25 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
Trade-offs
- Average benchmark score is below the index median; test it on your own task.
- Less reliable on open-ended reasoning and complex multi-step instructions.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Phi-3.5 1.5B | Microsoft | 32768 | $0.25 | 44.5 |
| SmolLM3 7B | Hugging Face | 16384 | $0.19 | 44.3 |
| StableLM 2 7B | Stability AI | 8192 | $0.28 | 44.3 |
| OLMo 2 1.5B | Allen Institute for AI | 16384 | $0.28 | 44.0 |
| SmolLM3 8B | Hugging Face | 32768 | $0.21 | 44.0 |
| GPT-NeoX 8B | EleutherAI | 16384 | $0.25 | 43.7 |
FAQ
How much does Phi-3.5 1.5B cost?
Phi-3.5 1.5B is listed at $0.25 per million input tokens and $1.05 per million output tokens. Cached input is $0.030 per million tokens. A blended 3:1 input-to-output mix is about $0.45 per million tokens.
What is the context window of Phi-3.5 1.5B?
Phi-3.5 1.5B has a context window of 32,768 tokens and can emit up to 4,096 output tokens per request in this index.
Is Phi-3.5 1.5B open weights?
Yes. Phi-3.5 1.5B is listed as an open-weight model under a open weights licence.
How does Phi-3.5 1.5B score on benchmarks?
Phi-3.5 1.5B averages 44.5 across the scored benchmarks in this index, ranking #582 of 600 models. Its strongest relative result is Arena Elo at 1145.1.
What is Phi-3.5 1.5B best for?
Phi-3.5 1.5B is a small model. That makes it a good fit for classification, extraction, routing, summarization and other high-volume jobs where latency and cost dominate. It is not the right pick for open-ended frontier reasoning, but at $0.25 per million input tokens it can handle a large share of production traffic.