Reasoning

Check exact model cardOpen weights where releasedUpdated August 2026

Phi-4

Phi-4 is a Phi-family model useful for small language model, edge, and low-resource workflow evaluation.

Microsoft · Phi

Editorial review

Reviewed byOpenSourcesAI EditorialLast updatedAugust 2026SourcesOpenSourcesAI checker catalog + Hugging Face model card (microsoft/phi-4)

Model checkpoints, context windows, provider support, local runtime compatibility, and license terms can change quickly. Verify the exact model card before production or commercial use.

Best for

Small language model reasoning and assistant workflows

Who should use it

  • Small language model reasoning and assistant workflows
  • Builders who want local or self-hosted testing options.

Common workflows

  • Small language model reasoning and assistant workflows
  • small workflows
  • edge workflows
  • local workflows
  • efficient workflows

Deployment and hardware notes

Local hardware needs vary by size, quantization, and runtime. Check the exact model card and serving stack.

License and usage notes

Check exact model card. Open weights where released. Verify the exact model card and license terms for the checkpoint or hosted provider you use.

Strengths

  • Open weights where released model option for Phi workflows.
  • Small language model reasoning and assistant workflows
  • Use the exact checkpoint and quantization that matches your hardware and latency target.

Limitations

  • Verify license, deployment requirements, runtime support, and fit on your own workload before production use.
  • Local hardware needs vary by size, quantization, and runtime. Check the exact model card and serving stack.
  • Context window and limits: 16,384 tokens.
  • Verify the exact model card, provider docs, license, and serving support before production use.

Local workflow notes

Use the exact checkpoint and quantization that matches your hardware and latency target.

Local runtimes: Ollama where supported, LM Studio where supported, llama.cpp where supported, Transformers

Platforms: Windows, macOS, Linux

Will Phi-4 run on your machine?

Phi-4 is 14B parameters and needs 10.6 GB of VRAM at Q4_K_M9.1 GB of weights plus 1.5 GB of runtime overhead for the inference server itself.

VRAM by quantization

QuantizationWeightsNeeds (with overhead)Quality
Q4_K_M9.1 GB10.6 GBgood
Q8_015 GB16.5 GBhigh
FP1628 GB29.5 GBreference

Fit on common hardware at Q4_K_M

HardwareMemory the model can useSystem RAMVerdict
CPU OnlyNone (CPU only)16 GBCPU offload
RTX 4060 Laptop8 GB16 GBCPU offload
RTX 3060 (12GB)12 GB32 GBTight
RTX 4060 Ti (16GB)16 GB32 GBComfortable
RTX 309024 GB64 GBComfortable
Apple Silicon (Unified Memory) 36 GB27 GB of 36 GB36 GBComfortable

Comfortable means VRAM clears the requirement by 2 GB or more. Tight means it covers the requirement with no margin. CPU offload means the model does not fit in VRAM but system RAM is at least 1.6× the weights, so it will run at reduced speed — expect roughly 1–5 tokens per second. Figures are weights plus a fixed runtime overhead and exclude KV-cache growth, which scales with context length.

Apple Silicon shares one pool of memory between the system and the GPU, so a model cannot use all of it. These rows apply the same 75% usable fraction the Compatibility Checker uses, which is why a 36 GB Mac is graded on less than 36 GB.

VRAM fit by quantization level

Enter your GPU VRAM below to see which quantization of Phi-4 fits and get the Ollama run command.

Related resources

Continue with model source notes, local tools, and implementation guides related to this model.

Hardware~9.1 GB at Q4_K_M (14B parameters)RuntimeOllama or LM Studio where supported, llama.cpp, Transformers, vLLMContext16,384 tokensLast updated2026
OpenSourcesAI checker catalog + Hugging Face model card (microsoft/phi-4)

Model ecosystem connections

Use these next-step links to move from this profile into related tools, comparisons, guides, stacks, and curated shortlists.