Fara1.5-4B Specialist

Released2026-05-21Size4.5B tensors · denseRoleSpecialistLicensemitContext256K config max

Where it runs 8K context · ≥ 5 t/s floor

Setup	Best plan	Memory	Generate
8 GB GPU + 32 GB RAM	native BF16vLLMGPU streams weights from RAM · 3 GiB in RAM	Fits*	≤ 10 t/s Ceiling
12 GB GPU + 32 GB RAM	native BF16vLLMFull GPU	Fits*	≤ 39 t/s Ceiling
16 GB GPU + 64 GB RAM	native BF16vLLMFull GPU	Fits*	≤ 97 t/s Ceiling
24 GB GPU + 64 GB RAM	native BF16vLLMFull GPU	Fits*	≤ 109 t/s Ceiling
32 GB GPU + 128 GB RAM	native BF16vLLMFull GPU	Fits*	≤ 195 t/s Ceiling
2× 24 GB GPUs + 128 GB RAM	native BF16vLLMFull GPU	Fits*	≤ 102 t/s Ceiling
96 GB workstation GPU + 256 GB RAM	native BF16vLLMFull GPU	Fits*	≤ 195 t/s Ceiling
Mac, 24 GB	Does not fit
Mac, 48 GB	Does not fit
Mac, 128 GB	Does not fit
Mac, 512 GB	Does not fit
DGX Spark 128 GB	native BF16vLLMUnified memory	Fits*	≤ 30 t/s Ceiling
Ryzen AI Max+ 395, 128 GB	native BF16vLLMUnified memory	Fits*	≤ 28 t/s Ceiling
CPU only, 64 GB DDR5-5600 dual channel	Does not fit

Same plan engine as the planner, with its default reserves and RAM. How plans are chosen.

Files exact Hugging Face bytes

Artifact	Size	Files	Revision	Source
Native safetensors BF16 microsoft/Fara1.5-4B	8.46 GiB	2	776a33ae5b	HF files

Not found in the representative quant search (not proven unavailable): GGUF FP16, GGUF Q8_0, GGUF Q5_K_M, GGUF Q4_K_M, MLX 8-bit, MLX 4-bit, EXL3, AWQ, GPTQ. Complete selected file sets; projectors, MTP/draft and lookup files are listed separately and budgeted only by plans that use them. File bytes are not allocated memory.

Quality

No Artificial Analysis evaluation mapped to this checkpoint.

Arena text rating

Unknown

Publisher-reported benchmarks

None extracted from the model card.

AA and Arena describe the hosted model in the stated mode, not each quantized file; scores are never inherited by base, pruned, abliterated or quantized variants and never averaged together.

How to run it

vLLM · unknown

publisher-documented, not executed source

vllm serve microsoft/Fara1.5-4B \
  --dtype bfloat16 \
  --max-model-len 262144 \
  --limit-mm-per-prompt image=10

Prerequisites: See source setup/dependencies; these snippets were inspected, not executed.

Computer-use task, not generic chat. Publisher recommends MagenticLite execution harness and its model-specific system prompt.
Image limit 10 and 256K context require multimodal memory; no fit claim.

Recipes are source examples (settings are the publisher’s or author’s, not measured optima) and were not executed by LocalFit. A downloadable file does not prove your backend build supports the architecture.

Reported speeds 0

No reported measurement yet. Submit one.