Model Follower.

Up to date

Rebuilt
2026-10-25 06:54 UTC
Release radar
117 on watch
Hub snapshot
2026-10-25
Re-checked
daily · every number sourced and dated

Models › phi-4 family

Phi-4-mini-instruct

Hugging Face ↗ Follow changes

microsoft MIT out 20 months ago

3.84B parameters, text in, text out, 128K tokens of context.

Verified 2026-10-25 Rank #178 in LLMs · score 90.5 method v1 Sources 3 Downloads 30d 345.5k · #202 of 619 tracked

Changed this week: 5 new quantized builds on the Hub (202 in all).2026-10-24 · 4 new quantized builds on the Hub (201 in all).2026-10-24

Can I run it?2.3GB at INT4Estimatefits a 16 GB cardmethodCommunity: Phi-4-mini-instruct — specs, VRAM, license, benchmarks | LocalLLMs →
Can I use it?MITPermissivelicence file2026-10-24
Do people use it?345.5kdownloads, 30 days#202 of 619 we track · top 33%+7 likes, downloads −7% this week868 likesHugging Face2026-10-24

Worth reading

Picked by rule from 25 tracked pieces: the lab's own post, the most shared, a field report, the latest.

Most discussed · 291 HN pts

Phi 4 available on Ollama

Phi-4 is a 14B parameter open model designed for memory/compute constrained environments and latency bound scenarios.

ollama.com 2025-01-09 291 HN pts · 135 comments 220 words about the phi 4 family family

25 best of 25 tracked

The 2 picks above, plus these 23. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.

  1. ArticlePhi-4: Microsoft's Newest Small Language Model Specializing in Complex Reasoningtechcommunity.microsoft.com · 2024-12-13 · 439 HN pts · official
  2. ArticlePhi-4 Reasoning Modelsazure.microsoft.com · 2025-05-01 · 131 HN pts · official
  3. ArticlePhi-4 Bug Fixesunsloth.ai · 2025-01-10 · 193 HN pts
  4. ArticleMicrosoft announces Phi-4-multimodal and Phi-4-miniazure.microsoft.com · 2025-02-26 · 48 HN pts · official
  5. ArticleIntroducing Phi-4-mini-flash-reasoningazure.microsoft.com · 2025-07-09 · 6 HN pts · official
  6. ArticlePhi-4-mini and Phi-4-multimodaltechcommunity.microsoft.com · 2025-03-01 · 5 HN pts · official
  7. ArticleMicrosoft Phi-4 reasoning technical report [pdf]microsoft.com · 2025-05-01 · 2 HN pts · official
  8. ArticlePhi-4 Technical Reportsimonwillison.net · 2024-12-16 · 5 HN pts
  9. ArticleMicrosoft debuts Phi-4, new generative AI modeltechcrunch.com · 2024-12-13
  10. PaperPhi-4 Technical Reportarxiv.org · 2024-12-13
  11. ArticleMicrosoft makes powerful Phi-4 model open-source on Hugging Faceventurebeat.com · 2025-01-09
  12. Articlemicrosoft / phi-4-mini-instructdocs.api.nvidia.com · undated
  13. Article[Literature Review] Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAsthemoonlight.io · undated
  14. ArticlePhi-4-mini-instruct — specs, VRAM, license, benchmarks | LocalLLMslocalllms.dev · undated
  15. ArticlePhi-4-mini Specs & Benchmarks | TheModelverse | The Modelversethemodelverse.in · undated
  16. ArticlePhi-4 Mini Instruct - Dataconomydataconomy.com · undated
  17. ArticlePhi-4 as LLM-as-Judge rivals specialized evaluator models wo/ finetuningflow-ai.com · 2025-01-22
  18. ArticleMicrosoft's smaller AI model beats the big guys: Meet Phi-4, the efficiency kingventurebeat.com · 2024-12-13
  19. ArticleReviewing Post-Training Techniques: DeepSeek, Qwen-2, and Phi-4brianfitzgerald.xyz · 2025-01-07
  20. ArticleNew for 2025 – more local models in Pieces, including Qwen Coder and Phi-4pieces.app · 2025-01-31
  21. ArticleMicrosoft's Phi-4-Mini: Never Has Small Been This Goodllmwatch.com · 2025-02-27
  22. ArticlePhi-4-mini-instruct (3.8B): Benchmarks & Specs | LLMLearnerllmlearner.com · 2025-02-27
  23. ArticlePhi-4 family — AI Stack Currentaistackcurrent.com · 2024-12-12

Where it's used

Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.

No API provider serves it, so its downloads are people running it themselves: 202 quantized builds and 100 Spaces.

  • As a service0no API provider listed
  • On own hardware202quantized builds
  • Built on340136 fine-tunes · 203 adapters · 1 merges
  • In apps100+Hugging Face Spaces

Top Spaces running it: pliny-the-prompter/obliteratus ♥ 621 · kevinwang676/F5-TTS ♥ 91 · mrfakename/DMOSpeech2 ♥ 19 · lfoppiano/document-qa ♥ 16

Will it fit?

Weights only, an estimate: leave room for the KV cache. Method

Fits a 16 GB card at INT4

About 2.3 GB at INT4, 4.6 GB at FP8, 9.2 GB at BF16: weights only, so leave room for the KV cache.

16 GB card64 GB Mac2×80 GB8×80 GBINT4: about 2.3 GB of weights (estimate; KV cache and activations extra)INT42.3 GBFP8: about 4.6 GB of weights (estimate; KV cache and activations extra)FP84.6 GBBF16: about 9.2 GB of weights (estimate; KV cache and activations extra)BF169.2 GB
16 GB card64 GB Mac2×80 GB8×80 GBINT4: about 2.3 GB of weights (estimate; KV cache and activations extra)INT42.3 GBFP8: about 4.6 GB of weights (estimate; KV cache and activations extra)FP84.6 GBBF16: about 9.2 GB of weights (estimate; KV cache and activations extra)BF169.2 GB

Estimate params × bytes per param × 1.2 · how we estimate

The story so far

Weights, every dated article, and repo commits on one rail.

Most coverage came in the first two weeks

Teal is the weights (2025-02-19), amber triangles are 10 dated articles (each links to it), grey ticks are repo commits.

JanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOct2025-12-10: 2 repo commits2025-05-01: 1 repo commit2025-03-10: 2 repo commits2025-03-05: 2 repo commits2025-03-02: 1 repo commit2025-03-01: 1 repo commit2025-02-28: 3 repo commits2025-02-27: 1 repo commit2025-02-26: 4 repo commits2025-02-25: 1 repo commit2025-02-24: 1 repo commit2025-02-19: 2 repo commitsPhi-4-reasoning-vision and the lessons of training a multimodal reasoning model · 2026-03-05Phi-4: Microsoft's Newest Small Language Model Specializing in Complex Reasoning · 2024-12-13Phi-4 Reasoning Models · 2025-05-01Phi-4 Bug Fixes · 2025-01-10Phi 4 available on Ollama · 2025-01-09Microsoft announces Phi-4-multimodal and Phi-4-mini · 2025-02-26Introducing Phi-4-mini-flash-reasoning · 2025-07-09Phi-4-mini and Phi-4-multimodal · 2025-03-01Microsoft Phi-4 reasoning technical report [pdf] · 2025-05-01Phi-4 Technical Report · 2024-12-16Weights 2025-02-19 · 2025-02-19Weights 2025-02-19now
JanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOct2025-12-10: 2 repo commits2025-05-01: 1 repo commit2025-03-10: 2 repo commits2025-03-05: 2 repo commits2025-03-02: 1 repo commit2025-03-01: 1 repo commit2025-02-28: 3 repo commits2025-02-27: 1 repo commit2025-02-26: 4 repo commits2025-02-25: 1 repo commit2025-02-24: 1 repo commit2025-02-19: 2 repo commitsPhi-4-reasoning-vision and the lessons of training a multimodal reasoning model · 2026-03-05Phi-4: Microsoft's Newest Small Language Model Specializing in Complex Reasoning · 2024-12-13Phi-4 Reasoning Models · 2025-05-01Phi-4 Bug Fixes · 2025-01-10Phi 4 available on Ollama · 2025-01-09Microsoft announces Phi-4-multimodal and Phi-4-mini · 2025-02-26Introducing Phi-4-mini-flash-reasoning · 2025-07-09Phi-4-mini and Phi-4-multimodal · 2025-03-01Microsoft Phi-4 reasoning technical report [pdf] · 2025-05-01Phi-4 Technical Report · 2024-12-16Weights 2025-02-19 · 2025-02-19Weights 2025-02-19now

What people built on it

542 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.

542 derivatives on the Hub, led by 203 adapters

From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.

202 quantized (Hugging Face model tree)136 fine-tunes (Hugging Face model tree)1 merge (Hugging Face model tree)203 adapters (Hugging Face model tree)202 quantized136 fine-tunes1 merge203 adapters
202 quantized (Hugging Face model tree)136 fine-tunes (Hugging Face model tree)1 merge (Hugging Face model tree)203 adapters (Hugging Face model tree)202 quantized136 fine-tunes1 merge203 adapters
GGUF MaziyarPanahi/Phi-4-mini-instruct-GGUF for llama.cpp, Ollama, LM Studio 3.9M downloads MLX mlx-community/Phi-4-mini-instruct-4bit for Apple silicon 73.2k downloads GPTQ iqbalamo93/Phi-4-mini-instruct-GPTQ-8bit for vLLM, SGLang 532k downloads

All variants →

Reference

Everything else, folded: the full tables, every source.

Benchmarks

18 rows

Benchmarks 18 rows

measured third-party leaderboard rowauthor-reported number from the model card or eval files (18)No confidence intervals unless the source gave them; settings differ between rows, so compare within a benchmark, not across. Date = the benchmark's own date when it gives one, else when we read it.
BenchmarkValueSetting / notesReported byDate
author-reported ARC Challenge · Chat and follow instructions83.7Phi-4 mini-Insmodel author2026-10-07
author-reported Arena-Hard · Chat and follow instructions32.8Phi-4 mini-Insmodel author2026-10-07
author-reported BBH · Reason through hard problems70.4Phi-4 mini-Insmodel author2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions81.2Phi-4 mini-Insmodel author2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions79.2Phi-4 mini-Insmodel author2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions77.6Phi-4 mini-Insmodel author2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions72.5Phi-4 mini-Insmodel author2026-10-07
author-reported GPQA Diamond · Reason through hard problems25.2Phi-4 mini-Insmodel author2026-10-07
author-reported GSM8K · Reason through hard problems88.6Phi-4 mini-Insmodel author2026-10-07
author-reported HellaSwag · Chat and follow instructions69.1Phi-4 mini-Insmodel author2026-10-07
author-reported MATH-500 · Reason through hard problems64Phi-4 mini-Insmodel author2026-10-07
author-reported MGSM · Work across languages63.9Phi-4 mini-Insmodel author2026-10-07
author-reported MMLU · Reason through hard problems67.3Phi-4 mini-Insmodel author2026-10-07
author-reported MMLU-Pro · Reason through hard problems52.8Phi-4 mini-Insmodel author2026-10-07
author-reported MMMLU · Work across languages49.3Phi-4 mini-Insmodel author2026-10-07
author-reported Overall63.5Phi-4 mini-Insmodel author2026-10-07
author-reported TruthfulQA · Chat and follow instructions66.4Phi-4 mini-Insmodel author2026-10-07
author-reported Winogrande · Chat and follow instructions67Phi-4 mini-Insmodel author2026-10-07

What the card says

3 cited sentences

Summary generated, cited

Phi-4-mini-instruct is a 3.84B parameter text-generation model from the Phi-4 family.[1][2] The model supports a 128K token context length and is released under an MIT license.[3][4] It is designed for multilingual commercial and research use in memory or compute constrained environments.[5]

  1. parameters human
  2. “Phi-4 model family”
  3. “supports 128K token context length”
  4. license
  5. “intended for broad multilingual commercial and research use”

Generated 2026-10-07 15:03 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.

Use-case evidence

7 use cases

What it is good at 7 use cases

Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.

Facts

22 fields

Listed in LLMs

Facts

Pipeline tagtext-generation
Librarytransformers
ArchitecturesPhi3ForCausalLM
Model typephi3
Dtypebfloat16
Params by dtypeBF16: 3,836,021,760
Size bandM
MoEno
Context length131,072 (config.json max_position_embeddings)
Languagesmultilingual, ar, zh, cs, da, nl, en, fi, fr, de, he, hu, it, ja, ko, no, pl, pt, ru, es, sv, th, tr, uk
PapersarXiv:2503.01743
LicenseMIT (mit)
Hub snapshot2026-10-25

Servability

vLLM · SGLang

Servability

vLLM: supported · Phi3ForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · Phi3ForCausalLM · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed

vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.

Hardware fit

3 estimates

Hardware fit ESTIMATE

Weights-only memory for Phi-4-mini-instruct: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.

BF16 (2 bytes/param)9.2 GB · Single GPU (<24 GB)
FP8 (1 byte/param)4.6 GB · Single GPU (<24 GB)
INT4 (0.5 bytes/param)2.3 GB · Edge (<4 GB)

Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.

All variants

40 variants

Lineage

Original model Family: microsoft/phi-4 542 derivatives on the Hub

quantized: 202finetune: 136adapter: 203merge: 1

Top variants (40 shown, ranked by downloads)

Relation
Quant
40 rows
Variant Relation Quant Params Band License Author Likes Downloads
MaziyarPanahi/Phi-4-mini-instruct-GGUF quantized GGUF — — — MaziyarPanahi 17 3.9M
unsloth/Phi-4-mini-instruct-GGUF quantized GGUF — — mit unsloth 159 709.2k
iqbalamo93/Phi-4-mini-instruct-GPTQ-8bit quantized GPTQ 3.84B M mit iqbalamo93 2 532k
bartowski/microsoft_Phi-4-mini-instruct-GGUF quantized GGUF — — — bartowski 52 461k
unsloth/Phi-4-mini-instruct-unsloth-bnb-4bit quantized -bnb 3.93B M mit unsloth 23 159.8k
unsloth/Phi-4-mini-instruct finetune — 3.84B M mit unsloth 22 150.6k
litert-community/Phi-4-mini-instruct finetune — — — mit litert-community 21 145.1k
Mungert/Phi-4-mini-instruct.gguf quantized gguf — — mit Mungert 28 86.2k
mlx-community/Phi-4-mini-instruct-4bit quantized -4bit 3.84B M mit mlx-community 5 73.2k
unsloth/Phi-4-mini-instruct-bnb-4bit quantized -bnb 3.94B M mit unsloth 7 64.5k
lmstudio-community/Phi-4-mini-instruct-GGUF quantized GGUF — — — lmstudio-community 17 61.6k
pytorch/Phi-4-mini-instruct-INT8-INT4 quantized INT8 — — mit pytorch 3 55.4k
pytorch/Phi-4-mini-instruct-FP8 quantized -FP8 — — mit pytorch 1 25.2k
OpenVINO/Phi-4-mini-instruct-int4-ov quantized int4 — — mit OpenVINO 6 16.5k
mlx-community/Phi-4-mini-instruct-8bit quantized -8bit 3.84B M mit mlx-community 7 15.7k
mmnga/Phi-4-mini-instruct-gguf quantized gguf — — mit mmnga 0 14.3k
DevQuasar-9/microsoft.Phi-4-mini-instruct-GGUF quantized GGUF — — — DevQuasar-9 2 13.9k
second-state/Phi-4-mini-instruct-GGUF quantized GGUF — — mit second-state 2 13.5k
gaianet/Phi-4-mini-instruct-GGUF quantized GGUF — — mit gaianet 0 12k
pytorch/Phi-4-mini-instruct-AWQ-INT4 quantized AWQ — — bsd-3-clause pytorch 4 11.5k
reallexi/lexi-coder-v4.3 adapter GGUF 3.84B M other reallexi 2 11.5k
arzaan789/phi-4-mini-uncensored quantized GGUF — — apache-2.0 arzaan789 4 9.5k
iqbalamo93/Phi-4-mini-instruct-GPTQ-4bit quantized GPTQ 3.84B M mit iqbalamo93 0 8k
pytorch/Phi-4-mini-instruct-INT4 quantized INT4 — — bsd-3-clause pytorch 1 7.7k
matrixportalx/Phi-4-mini-instruct-Q4_K_M-GGUF quantized GGUF — — mit matrixportalx 4 7k
DuoNeural/Phi-4-mini-instruct-GGUF quantized GGUF — — mit DuoNeural 1 7k
robertou2/task-13-microsoft-Phi-4-mini-instruct adapter — — — — robertou2 0 5.3k
flockgo/task-13-microsoft-Phi-4-mini-instruct adapter — — — — flockgo 0 5.2k
lfhe/FLock-Arena-Task-17-JustParse adapter — — — — lfhe 0 5.1k
RedHatAI/Phi-4-mini-instruct-FP8-dynamic quantized -FP8 4.45B M mit RedHatAI 1 4.9k
flockgo/task-17-microsoft-Phi-4 adapter — — — — flockgo 0 4.8k
ariel-pillar/phi-4_function_calling quantized GGUF — — — ariel-pillar 0 4.3k
tensorblock/Phi-4-mini-instruct-GGUF quantized GGUF — — mit tensorblock 0 4k
AhtnaGlen/phi-4-mini-instruct-int4-sym-npu-ov finetune int4 — — mit AhtnaGlen 0 3.6k
robertou2/task-17-microsoft-Phi-4-mini-instruct adapter — — — — robertou2 0 3.3k
jerryzh168/phi4-mini-int4wo-gemlite quantized — — — mit jerryzh168 0 3.2k
AtomicChat/Phi-4-mini-instruct-GGUF quantized GGUF — — mit AtomicChat 2 2.5k
tatsuyaaaaaaa/Phi-4-mini-instruct-gguf quantized gguf — — mit tatsuyaaaaaaa 0 2.4k
DreamGallery/task-13-microsoft-Phi-4-mini-instruct adapter — — — — DreamGallery 0 2.2k
Josephgflowers/Phinance-Phi-4-mini-instruct-finance-v0.4-with-reasoning finetune — 3.84B M mit Josephgflowers 2 2.1k

Repository changelog

21 commits
changed in the last 7 daysOther: 2026-10-24, 2026-10-24all changes

Repository changelog 21 commits

DateCommitSHA
2025-12-10 Update README.md (#43) cfbefac
2025-12-10 Upload data_summary_card.md (#42) 7c44507
2025-05-01 Update urls 5a14955
2025-03-10 Add metadata and links to model card (#16) c0fb9e7
2025-03-10 Update sample_finetune.py (#17) fad2462
2025-03-05 Update readme d02e859
2025-03-05 Update readme 5889daf
2025-03-02 Update readme b93bc76
2025-03-01 Update README.md (#10) 2ebeb38
2025-02-28 Update README.md (#9) 0fd084b
2025-02-28 Fix "Norway problem" in YAML (#7) f7a0a61
2025-02-28 Add language tag (#5) f984c15
2025-02-27 Update readme 4b00ec8
2025-02-26 Update readme 12cce9e
2025-02-26 update readme 4665bce
2025-02-26 update blog url 4fcd2e8
2025-02-26 Update README 66561c5
2025-02-25 update readme a58b82a
2025-02-24 Update readme f8b7731
2025-02-19 Added model files 618e63e
2025-02-19 initial commit fde0620

Articles and coverage

25 pieces

Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency. Summaries are generated from a stored copy of the article text; every sentence cites a span of that text, and nothing is shown that failed the citation check.

Official7

Phi-4-reasoning-vision and the lessons of training a multimodal reasoning model
microsoft.com2026-03-05officialauthorityabout the phi 4 familyHN 93 pts · 11 comments4,049 words
Phi-4-reasoning-vision-15B is a 15 billion parameter open-weight multimodal reasoning model trained with just 200 billion tokens of multimodal data. The model excels at math and science reasoning and understanding user-interfaces, presenting an appealing value relative to popular open-weight models.generated, cited
Phi-4: Microsoft's Newest Small Language Model Specializing in Complex Reasoning
techcommunity.microsoft.com2024-12-13officialauthorityHN front pageabout the phi 4 familyHN 439 pts · 143 comments5,664 words
Microsoft introduces Phi-4, a 14B parameter small language model designed for complex reasoning in math and language tasks. The model achieves performance through the use of high-quality synthetic datasets, curation of high-quality organic data, and post-training innovations. Phi-4 outperforms much larger models, including Gemini Pro 1.5, on math competition problems.written, reviewed, cited
  • Phi-4 is available on Azure AI Foundry and Hugging Face as of the Dec 12, 2024 announcement.
  • Developers can integrate content safety features into applications easily through a single API.
  • The blog post was updated on Jan 08, 2025, indicating potential changes after the initial release.
Phi-4 Reasoning Models
azure.microsoft.com2025-05-01officialauthorityHN front pageabout the phi 4 familyHN 131 pts · 26 comments1,007 words
Microsoft releases Phi-4-reasoning, a 14-billion parameter open-weight model, alongside Phi-4-reasoning-plus and Phi-4-mini-reasoning on Azure AI Foundry and HuggingFace. Phi-4-reasoning-plus utilizes 1.5x more tokens than Phi-4-reasoning to deliver higher accuracy, while both models outperform the 671-billion parameter DeepSeek-R1 on the AIME 2025 test. Optimized versions using ONNX are now available for Snapdragon powered Copilot+ PCs to run efficiently on NPUs.written, reviewed, cited
  • Phi-4-reasoning is trained via supervised fine-tuning of Phi-4 on carefully curated reasoning demonstrations from OpenAI o3-mini.
  • Phi-4-mini-reasoning is fine-tuned with synthetic data generated by Deepseek-R1 model to balance efficiency with advanced reasoning ability.
  • The NPU-optimized Phi Silica variant is designed to be preloaded in memory for blazing fast time to first token responses.
Microsoft announces Phi-4-multimodal and Phi-4-mini
azure.microsoft.com2025-02-26officialauthorityabout the phi 4 familyHN 48 pts · 3 comments2,100 words
Introducing Phi-4-mini-flash-reasoning
azure.microsoft.com2025-07-09officialauthorityabout the phi 4 familyHN 6 pts · 1 comments847 words
Phi-4-mini and Phi-4-multimodal
techcommunity.microsoft.com2025-03-01officialauthorityabout the phi 4 familyHN 5 pts · 1 comments5,908 words
Microsoft Phi-4 reasoning technical report [pdf]
microsoft.com2025-05-01officialauthorityabout the phi 4 familyHN 2 pts · 1 comments

Analysis3

Phi-4 Bug Fixes
unsloth.ai2025-01-10authorityHN front pageabout the phi 4 familyHN 193 pts · 68 comments
Phi 4 available on Ollama
ollama.com2025-01-09HN front pageabout the phi 4 familyHN 291 pts · 135 comments220 words
Phi-4 is a 14B parameter open model designed for memory/compute constrained environments and latency bound scenarios. It supports a context length of 16k tokens and is specifically optimized for reasoning and logic tasks. The model focuses primarily on English, requiring developers to evaluate accuracy and safety before deployment in high-risk scenarios.written, reviewed, cited
Phi-4 Technical Report
simonwillison.net2024-12-16authorityabout the phi 4 familyHN 5 pts · 0 comments740 words
All coverage tracked for this model (25) — append-only, nothing is ever removed
TitleSourcePublishedFirst seenSignals
microsoft / phi-4-mini-instructdocs.api.nvidia.com—2026-10-07exa
[Literature Review] Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAsthemoonlight.io—2026-10-07exa
Phi-4-mini-instruct — specs, VRAM, license, benchmarks | LocalLLMslocalllms.dev—2026-10-07exa
Phi-4-mini Specs & Benchmarks | TheModelverse | The Modelversethemodelverse.in—2026-10-07exa
Phi-4 Mini Instruct - Dataconomydataconomy.com—2026-10-07exa
Phi-4-reasoning-vision and the lessons of training a multimodal reasoning modelmicrosoft.com official2026-03-052026-10-07hn
Introducing Phi-4-mini-flash-reasoningazure.microsoft.com official2025-07-092026-10-07hn
Microsoft Phi-4 reasoning technical report [pdf]microsoft.com official2025-05-012026-10-07hn
Phi-4 Reasoning Modelsazure.microsoft.com official2025-05-012026-10-07hn
Phi-4-mini and Phi-4-multimodaltechcommunity.microsoft.com official2025-03-012026-10-07hn
Microsoft's Phi-4-Mini: Never Has Small Been This Goodllmwatch.com2025-02-272026-10-07exa
Phi-4-mini-instruct (3.8B): Benchmarks & Specs | LLMLearnerllmlearner.com2025-02-272026-10-07exa
Microsoft announces Phi-4-multimodal and Phi-4-miniazure.microsoft.com official2025-02-262026-10-07hn
New for 2025 – more local models in Pieces, including Qwen Coder and Phi-4pieces.app2025-01-312026-10-07hn
Phi-4 as LLM-as-Judge rivals specialized evaluator models wo/ finetuningflow-ai.com2025-01-222026-10-07hn
Phi-4 Bug Fixesunsloth.ai2025-01-102026-10-07hn
Phi 4 available on Ollamaollama.com2025-01-092026-10-07hn
Microsoft makes powerful Phi-4 model open-source on Hugging Faceventurebeat.com2025-01-092026-10-07hn
Reviewing Post-Training Techniques: DeepSeek, Qwen-2, and Phi-4brianfitzgerald.xyz2025-01-072026-10-07hn
Phi-4 Technical Reportsimonwillison.net2024-12-162026-10-07hn
Phi-4 Technical Reportarxiv.org2024-12-132026-10-07hn
Microsoft's smaller AI model beats the big guys: Meet Phi-4, the efficiency kingventurebeat.com2024-12-132026-10-07hn
Microsoft debuts Phi-4, new generative AI modeltechcrunch.com2024-12-132026-10-07hn
Phi-4: Microsoft's Newest Small Language Model Specializing in Complex Reasoningtechcommunity.microsoft.com official2024-12-132026-10-07hn
Phi-4 family — AI Stack Currentaistackcurrent.com2024-12-122026-10-07exa

Papers

1 paper

Papers 1

arXiv links found in the model card and its tags. "Peer-reviewed" is shown only with evidence: a published venue in OpenAlex (measured) or the authors' own journal reference on arXiv (reported).

Similar models

3 models

Similar models

  • Qwen3-4B · 4.02B
    same size band (M), same task (text-generation), more providers, permissive license
  • Qwen3-4B-Instruct-2507 · 4.02B
    same size band (M), same task (text-generation), more providers, permissive license
  • Phi-3-mini-4k-instruct · 3.82B
    same size band (M), same task (text-generation), more providers, permissive license

Sources

3 sources

Sources

Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:35 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:45 UTC · Summary: 2026-10-07 15:03 UTC · Last verified 2026-10-25

Follow Phi-4-mini-instruct

Watch this model

Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.