Model

Phi-3.5-mini (3.8B)

OPEN

Microsoft · Phi-3.5 family · released Aug 20, 2024

Refresh of Phi-3-mini with 22-language multilingual support and 128K context; beats Llama-3.1 8B on many tasks.

ReasoningCodingVisionFunction callingTool useAgentic

477.1

Elo · rank #313

Parameters

3.8B

Active params

3.8B (dense)

Context

128K tokens

Architecture

Dense decoder-only Transformer

License

MIT

Languages

22+

API price (in/out)

$0.13 / $0.52

Modalities

text

Benchmark results

Bar shows position within the tracked field; marker = field best

ARC-ChallengeReasoning84.6%#32

best: Llama 3.1 405B · 96.9%

Arena-HardHuman preference37.0%#48

best: Qwen3-235B-A22B (Non-Thinking) · 96.1%

BIG-Bench HardReasoning69.0%#62

best: ERNIE 4.5 300B-A47B · 94.3%

GPQA DiamondReasoning30.4%#260

best: GPT-5.6 · 94.6%

GSM8KMath86.2%#61

best: Llama 3.1 405B · 96.8%

HellaSwagReasoning69.4%#126

best: Claude 3 Opus · 95.4%

HumanEvalCoding62.8%#96

best: Claude Opus 4.5 · 99.4%

IFEvalReasoning57.7%#125

best: Gemma 4 26B A4B · 98.5%

MATH-500Math48.5%#137

best: GPT-5 · 99.4%

MBPPCoding69.6%#41

best: Llama-3.3-Nemotron-Super-49B v1 (Reasoning On) · 91.3%

MGSMMath47.9%#47

best: OpenAI o4-mini · 93.7%

MMLU-ProKnowledge47.4%#137

best: Claude Fable 5 · 91.5%

MMLUKnowledge69.0%#153

best: OpenAI o3 · 92.9%

MMMLUKnowledge55.4%#52

best: Gemini 3.1 Pro · 92.6%

PIQAReasoning81.0%#36

best: GPT-4o mini · 93.1%

Social IQaReasoning74.7%#8

best: Apple DCLM-Baseline 7B · 82.9%

TruthfulQAKnowledge64.0%#19

best: Phi-3.5-MoE (16x3.8B, 6.6B active) · 77.5%

WinoGrandeReasoning68.5%#107

best: PaLM 2 · 90.9%

Run it locally

VRAM @ Q4

4 GB

VRAM @ FP16

8 GB

Fits on (Q4)

RTX 3060 12GBRTX 4070 Ti 16GBRTX 3090 24GBRTX 4090 24GBRTX 5090 32GBM4 Pro 48GBM3 Max 128GBM3 Ultra 512GBA100 80GBH100 80GBH200 141GBB200 192GB

Throughput data unavailable.

Quantizations

GGUF Q4 · AWQ · ONNX INT4 · MLX 4-bit

Phi-3.5 family

Elo progression across releases

Phi-3.5-mini (3.8B)Aug 2024477.1 Phi-3.5-MoE (16x3.8B, 6.6B active)Aug 2024791.5 Phi-3.5-vision (4.2B)Aug 2024532.4

Resources

HFmicrosoft/Phi-3.5-mini-instruct GHmicrosoft/PhiCookBook DOCStechcommunity.microsoft.com/blog/azure-ai-foundry-blog/discover-the-new-multi-lingual-high-quality-phi-3-5-slms/4225280 APIAvailable via Microsoft API

API price $0.13/$0.52 · each benchmark row carries its own source badge (see methodology)