๐ Microsoft
Microsoft
Phi 3.5 Mini 4B
Legacy1.2MDownloads993LikesAug 2024Released128K tokensContextMITLicense39 BasicQuality
Phi 3.5 Mini 4B (4B parameters) requires approximately 10.1 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 12 GB of VRAM.
Get started
โ copy & paste to run locallyCopy-paste commands to run Phi 3.5 Mini 4B on your machine.
Run
ollama run phi3.5Quick specs
Parameters4B
Architecturedense
Context128K tokens
Modalitytext
Min RAM1.6 GB
Rec. RAM2.4 GB (Q4_K_M)
LicenseMIT
FamilyPhi
โ Chat
About this model
- โขMemory/compute constrained environments
- โขLatency bound scenarios
- โขStrong reasoning (especially code, math and logic)
Related models
Your hardware
Detecting...
Quick picks
๐ Intel
๐ NVIDIA
Best budgetB
Intel Arc B580 12GB~$249 โ 56 tok/sBest overallA
RTX 5070 Ti 16GB~$749 โ 76 tok/sBest hardware
Top picks for Phi 3.5 Mini 4B
Run this model
Quantization options
VRAM estimates by quant level
No hardware detected โ fit column shows raw VRAM estimates
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 1.6 GB | Low | โ |
Q3_K_S | 3 | 2.0 GB | Low | โ |
NVFP4 | 4 | 2.2 GB | Medium | โ |
Q4_K_M | 4 | 2.4 GB | Medium | โ |
Q5_K_M | 5 | 2.9 GB | High | โ |
Q6_K | 6 | 3.3 GB | High | โ |
Q8_0 | 8 | 4.3 GB | Very High | โ |
F16 | 16 | 8.2 GB | Maximum | โ |
Quality benchmarks
Phi 3.5 Mini 4B benchmark scores
Coding
SWE-bench Verifiedโ
HumanEval+62.8%
Aider Polyglotโ
LiveCodeBenchโ
Reasoning
MMLU-Pro69.0%
GPQA Diamond12.0%
MATH-50019.6%
ARC Challenge84.6%
General
Chatbot Arenaโ
IFEval57.7%
Source: official ยท 2024-08-20
Hardware compatibility
Fit estimates across all hardware
Computing compatibility...
Memory breakdown
Reference: RTX 2060 6GB
Weights2.4 GB
KV Cache5.9 GB
Runtime1.2 GB
Headroom0.6 GB
Frequently asked questions
FAQ โ Phi 3.5 Mini 4B
See also
