Model

Phi-1 (1.3B)

OPEN

Microsoft · Phi-1 family · released Jun 20, 2023

Launched 'Textbooks Are All You Need'; 1.3B code model hitting ~50.6% HumanEval via curated synthetic data over scale.

ReasoningCodingVisionFunction callingTool useAgentic

328.3

Elo · unrated

Parameters

1.3B

Active params

1.3B (dense)

Context

2K tokens

Architecture

Dense Transformer decoder

License

MIT

Languages

—

API price (in/out)

No hosted API

Modalities

text

Benchmark results

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

HumanEvalCoding50.6%#123

best: Claude Opus 4.5 · 99.4%

MBPPCoding55.5%#64

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

Run it locally

VRAM @ Q4

1 GB

VRAM @ FP16

3 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

Fine-tune it

Permissive

QLoRA2.8 GB1× RTX 3060 12GB

LoRA4.7 GB1× RTX 3060 12GB

Full fine-tune22.8 GB1× RTX 3090 24GB

QLoRA SFT on ~10k samples ≈ $0.76 (1× RTX 3060 12GB)

Plan a fine-tune →

Phi-1 family

Elo progression across releases

Phi-1 (1.3B)Jun 2023328.3

Resources

HFhttps://huggingface.co/microsoft/phi-1

API price weights · each benchmark row carries its own source badge (see methodology)