Model datasheet · Phi

Phi-4

The exam ace with no life experience: superb STEM reasoning per GB, near-zero world knowledge, and only a 16K window. Excellent at closed problems you paste in; pair with retrieval for everything else.

Vendor: Microsoft
Architecture: Dense · 14B
Context: 16,384 tokens
License: MIT
Released: 2024-12-12
Vision: No

→ GGUF (llama.cpp / LM Studio / Ollama) → MLX (Apple Silicon)

§1 Characteristics by quantization band

Q2–Q3 — 2–3 bit

7GB weights · e.g. Q3_K_M + KV: 1.6GB at 8K · 3.1GB at 32K · 3.1GB at 128K

— Not yet rated · contributions welcome —

Q4–Q5 — 4–5 bit

9GB weights · e.g. Q4_K_M , mlx-4bit + KV: 1.6GB at 8K · 3.1GB at 32K · 3.1GB at 128K

Math & reasoning

6/10

data checked Aug 2026

quant-degradation-community — KLD/perplexity delta vs bf16 @ Q4_K_M · aggregated · 2026-08-30
Editorial derivation: small loss at 4-bit.

Coding

6/10

data checked Aug 2026

quant-degradation-community — KLD/perplexity delta vs bf16 @ Q4_K_M · aggregated · 2026-08-30
Editorial derivation: small loss at 4-bit.

Q6–Q8 — 6–8 bit

12GB weights · e.g. Q6_K , Q8_0 + KV: 1.6GB at 8K · 3.1GB at 32K · 3.1GB at 128K

— Not yet rated · contributions welcome —

FP16 — full precision (fp16/bf16)

29GB weights · e.g. bf16 + KV: 1.6GB at 8K · 3.1GB at 32K · 3.1GB at 128K

Math & reasoning

7/10

data checked Aug 2026

vendor-model-card — MATH = 80.4 @ bf16 · vendor-reported · 2026-08-30
MGSM: 80.6, GPQA: 56.1 — exceptional per-GB math.

Coding

7/10

data checked Aug 2026

vendor-model-card — HumanEval = 82.6 @ bf16 · vendor-reported · 2026-08-30
Absent from the Aider polyglot leaderboard as of 2026-08-30.

Factuality

5/10

data checked Aug 2026

vectara-hallucination — HHEM hallucination rate % (lower is better) = 3.7 @ bf16 · aggregated · 2026-08-30
Grounds provided text very well.
vendor-model-card — SimpleQA = 3 @ bf16 · vendor-reported · 2026-08-30
…but knows almost nothing: open-domain recall is among the lowest measured. The split score reflects both.

§2 Known issues & what fixes them

ISSUE-01 severe Q2–Q3 / Q4–Q5 / Q6–Q8 / FP16 · factuality

Near-zero open-domain knowledge (SimpleQA 3.0 — vendor's own number). It will confidently invent facts it was never taught, even though it grounds provided text well (HHEM 3.7%).

Fix

RAG (retrieval) [strong]

Feed it the facts and it behaves: its grounding is excellent. Never use it as an encyclopedia.

Fix

MCP / external tools [strong]

Web-search tooling covers the recall gap for current-events questions.

Evidence · 2 sources · community-consensus

Phi-4 model card (SimpleQA 3.0, self-reported) (vendor, 2026-08-30)
Vectara HHEM leaderboard (grounding vs recall split) (leaderboard, 2026-08-30)

ISSUE-02 moderate Q2–Q3 / Q4–Q5 / Q6–Q8 / FP16 · long-context

16K context window — a quarter of what 2025+ peers offer. Long documents and long agent sessions simply don't fit.

Fix

RAG (retrieval) [partial]

Chunked retrieval keeps each request inside 16K.

No fix

No known fix — choose a different model

For genuinely long context, pick a 128K model.

Evidence · 1 source · anecdotal

Phi-4 model card (16K context) (vendor, 2026-08-30)

ISSUE-03 moderate Q2–Q3 / Q4–Q5 / Q6–Q8 / FP16 · instruction-following

Benchmark scores flatter it: in longer interactive sessions it drifts off instructions and feels less capable than its numbers suggest.

Fix

Prompt technique [partial]

Use it for one-shot, well-specified tasks; restate constraints each turn in longer sessions.

Evidence · 1 source · anecdotal

Tried Phi-4 — it didn't impress (interactive-use writeup) (blog, 2026-08-30)