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 › gemma-4 family

gemma-4-E4B-it

Hugging Face ↗ Follow changes

google Apache 2.0 out 8 months ago

8.00B parameters, several kinds of input and output, 128K tokens of context.

Verified 2026-10-25 Rank #51 in LLMs · score 107.5 method v1 Sources 3 Downloads 30d 4M · #36 of 619 tracked

Changed this week: 27 new quantized builds on the Hub (416 in all).2026-10-24 · 17 new quantized builds on the Hub (406 in all).2026-10-24

Can I run it?4.8GB at INT4Estimatefits a 16 GB cardmethod
Can I use it?Apache 2.0Permissive; patent grantlicence file2026-10-24
Do people use it?4Mdownloads, 30 days#36 of 619 we track · top 6%+39 likes, downloads −5% this week1,770 likesHugging Face2026-10-24

Worth reading

Picked by rule from 37 tracked pieces: the lab's own post, the most shared, a field report, the latest.

Most discussed · 868 HN pts

Gemma 4 on iPhone

apps.apple.com 2026-04-05 868 HN pts · 234 comments about the gemma 4 family family

25 best of 37 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. CodeShow HN: Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Macgithub.com · 2026-07-29
  2. ArticleGemma 4 12B: A unified, encoder-free multimodal modelblog.google · 2026-06-03 · official
  3. ArticleRunning Gemma 4 26B at 5 tokens/sec on a 13-year-old Xeon with no GPUneomindlabs.com · 2026-07-15
  4. CodeShow HN: Cactus Hybrid: We taught Gemma 4 to know when it's wronggithub.com · 2026-07-22 · 191 HN pts
  5. ArticleGemma 4 QAT models: Optimizing compression for mobile and laptop efficiencyblog.google · 2026-06-05 · 406 HN pts · official
  6. ArticleAccelerating Gemma 4: faster inference with multi-token prediction draftersblog.google · 2026-05-05 · 687 HN pts · official
  7. ArticleGoogle releases Gemma 4 open modelsdeepmind.google · 2026-04-02 · 1,812 HN pts · official
  8. ArticleFable 5 pushed Gemma 4 to 255 tok/s on WebGPUxcancel.com · 2026-06-18 · 48 HN pts
  9. CodeGemma 4 E4B running inside Obsidian via WebGPU and Karpathy-style LLM Wikigithub.com · 2026-09-09 · 4 HN pts
  10. ArticleGoogle Gemma 4 Runs Natively on iPhone with Full Offline AI Inferencegizmoweek.com · 2026-04-15 · 303 HN pts
  11. CodeApril 2026 TLDR Setup for Ollama and Gemma 4 26B on a Mac minigist.github.com · 2026-04-03
  12. ArticleShow HN: Determinstic LLM inference for lowest price Gemma 4, with Windows XPtokendelivery.ai · 2026-09-12 · 9 HN pts
  13. ArticleI ran Gemma 4 as a local model in Codex CLIblog.danielvaughan.com · 2026-04-12
  14. ArticleRunning Gemma 4 locally with LM Studio's new headless CLI and Claude Codeai.georgeliu.com · 2026-04-05
  15. ArticleGemma 4 on Cerebras - The Fastest Inference Is Now Multimodalcerebras.ai · 2026-06-30
  16. CodeShow HN: Gemma 4 Multimodal Fine-Tuner for Apple Silicongithub.com · 2026-04-07
  17. ArticleShow HN: Prompt-to-Excalidraw demo with Gemma 4 E2B in the browser (3.1GB)teamchong.github.io · 2026-04-19
  18. CodeShow HN: Local privacy-first Microsoft Recall alternative with Gemma 4github.com · 2026-07-04
  19. ArticleGemma 4: Byte for byte, the most capable open modelsblog.google · 2026-04-02 · official
  20. ArticleA Visual Guide to Gemma 4 12Bnewsletter.maartengrootendorst.com · 2026-06-03
  21. CodeShow HN: Local LLM code-generation with Gemma 4 e2B via JSON AST to Clojuregithub.com · 2026-05-19
  22. ArticleShow HN: Running Gemma-4 26B at 124 tokens/SEC on a CPU, no GPUapeg.dev · 2026-06-30
  23. ArticleShow HN: I made a Gemma 4 Mac app that names screenshots with local AIsnapname.app · 2026-05-31

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: 416 quantized builds and 100 Spaces.

  • As a service0no API provider listed
  • On own hardware416quantized builds
  • Built on850427 fine-tunes · 420 adapters · 3 merges
  • In apps100+Hugging Face Spaces

Top Spaces running it: embedl/hfviewer ♥ 181 · huggingface-projects/gemma-4-e4b-it ♥ 38 · davanstrien/diffusiongemma-ocr-correction ♥ 23 · FINAL-Bench/Darwin-4B-Opus ♥ 11

Will it fit?

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

Fits a 16 GB card at INT4

About 4.8 GB at INT4, 9.6 GB at FP8, 19 GB at BF16: weights only, so leave room for the KV cache.

16 GB card64 GB Mac2×80 GB8×80 GBINT4: about 4.8 GB of weights (estimate; KV cache and activations extra)INT44.8 GBFP8: about 9.6 GB of weights (estimate; KV cache and activations extra)FP89.6 GBBF16: about 19 GB of weights (estimate; KV cache and activations extra)BF1619 GB
16 GB card64 GB Mac2×80 GB8×80 GBINT4: about 4.8 GB of weights (estimate; KV cache and activations extra)INT44.8 GBFP8: about 9.6 GB of weights (estimate; KV cache and activations extra)FP89.6 GBBF16: about 19 GB of weights (estimate; KV cache and activations extra)BF1619 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.

10 dated articles and 12 repo commits so far

Teal is the weights (2026-03-02), amber triangles are 10 dated articles (each links to it), grey ticks are repo commits.

AprMayJunJulAugSepOct2026-07-20: 1 repo commit2026-07-15: 2 repo commits2026-07-10: 1 repo commit2026-06-03: 1 repo commit2026-06-02: 1 repo commit2026-05-18: 1 repo commit2026-05-07: 1 repo commit2026-04-28: 1 repo commit2026-04-10: 2 repo commits2026-04-02: 1 repo commitShow HN: Cactus Hybrid: We taught Gemma 4 to know when it's wrong · 2026-07-22Gemma 4 QAT models: Optimizing compression for mobile and laptop efficiency · 2026-06-05Accelerating Gemma 4: faster inference with multi-token prediction drafters · 2026-05-05Google releases Gemma 4 open models · 2026-04-02Gemma 4 E2B inference in 700 lines of C · 2026-08-27Fable 5 pushed Gemma 4 to 255 tok/s on WebGPU · 2026-06-18Gemma 4 on iPhone · 2026-04-05Gemma 4 E4B running inside Obsidian via WebGPU and Karpathy-style LLM Wiki · 2026-09-09Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference · 2026-04-15Show HN: Determinstic LLM inference for lowest price Gemma 4, with Windows XP · 2026-09-12Weights 2026-03-02 · 2026-03-02Weights 2026-03-02now
AprMayJunJulAugSepOct2026-07-20: 1 repo commit2026-07-15: 2 repo commits2026-07-10: 1 repo commit2026-06-03: 1 repo commit2026-06-02: 1 repo commit2026-05-18: 1 repo commit2026-05-07: 1 repo commit2026-04-28: 1 repo commit2026-04-10: 2 repo commits2026-04-02: 1 repo commitShow HN: Cactus Hybrid: We taught Gemma 4 to know when it's wrong · 2026-07-22Gemma 4 QAT models: Optimizing compression for mobile and laptop efficiency · 2026-06-05Accelerating Gemma 4: faster inference with multi-token prediction drafters · 2026-05-05Google releases Gemma 4 open models · 2026-04-02Gemma 4 E2B inference in 700 lines of C · 2026-08-27Fable 5 pushed Gemma 4 to 255 tok/s on WebGPU · 2026-06-18Gemma 4 on iPhone · 2026-04-05Gemma 4 E4B running inside Obsidian via WebGPU and Karpathy-style LLM Wiki · 2026-09-09Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference · 2026-04-15Show HN: Determinstic LLM inference for lowest price Gemma 4, with Windows XP · 2026-09-12Weights 2026-03-02 · 2026-03-02Weights 2026-03-02now

What people built on it

1,266 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.

1,266 derivatives on the Hub, led by 427 fine-tunes

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

416 quantized (Hugging Face model tree)427 fine-tunes (Hugging Face model tree)3 merges (Hugging Face model tree)420 adapters (Hugging Face model tree)416 quantized427 fine-tunes3 merges420 adapters
416 quantized (Hugging Face model tree)427 fine-tunes (Hugging Face model tree)3 merges (Hugging Face model tree)420 adapters (Hugging Face model tree)416 quantized427 fine-tunes3 merges420 adapters
GGUF unsloth/gemma-4-E4B-it-GGUF for llama.cpp, Ollama, LM Studio 6M downloads MLX lmstudio-community/gemma-4-E4B-it-MLX-4bit for Apple silicon 6.9M downloads AWQ cyankiwi/gemma-4-E4B-it-AWQ-INT4 for vLLM, SGLang 958.6k downloads

All variants →

Reference

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

Benchmarks

14 rows

Benchmarks 14 rows

measured third-party leaderboard rowauthor-reported number from the model card or eval files (14)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 AIME 2026 · Reason through hard problems42.5% percentGemma 4 E4Bmodel author2026-10-07
author-reported BBEH · Reason through hard problems33.1% percentGemma 4 E4Bmodel author2026-10-07
author-reported CoVoST35.54Gemma 4 E4Bmodel author2026-10-07
author-reported Codeforces · Write code with agents940 eloGemma 4 E4Bmodel author2026-10-07
author-reported FLEURS (lower is better)0.08 otherGemma 4 E4Bmodel author2026-10-07
author-reported GPQA Diamond · Reason through hard problems58.6% percentGemma 4 E4Bmodel author2026-10-07
author-reported LiveCodeBench · Write code with agents52.0% percentGemma 4 E4Bmodel author2026-10-07
author-reported MMLU-Pro · Reason through hard problems69.4% percentGemma 4 E4Bmodel author2026-10-07
author-reported MMMLU · Work across languages76.6% percentGemma 4 E4Bmodel author2026-10-07
author-reported MMMU-Pro · Understand images and video52.6% percentGemma 4 E4Bmodel author2026-10-07
author-reported MathVision · Understand images and video59.5% percentGemma 4 E4Bmodel author2026-10-07
author-reported MedXPertQA MM28.7% percentGemma 4 E4Bmodel author2026-10-07
author-reported OmniDocBench · Extract text from documents0.18 otherGemma 4 E4Bmodel author2026-10-07
author-reported tau-bench · Call tools and APIs42.2% percentGemma 4 E4Bmodel author2026-10-07

What the card says

1 cited sentence

Summary generated, cited

The model supports text, image, and audio modalities and is released under the Apache 2.0 license.[1][2]

  1. “Text, Image, Audio”
  2. license

Generated 2026-10-07 15:00 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.

Use-case evidence

9 use cases

What it is good at 9 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

21 fields

Listed in LLMs

Facts

Pipeline tagany-to-any
Librarytransformers
ArchitecturesGemma4ForConditionalGeneration
Model typegemma4
Dtypebfloat16
Params by dtypeBF16: 7,996,156,490
Size bandM
MoEno
Context length131,072 (config.json max_position_embeddings)
PapersarXiv:2607.02770
LicenseApache 2.0 (apache-2.0)
Hub snapshot2026-10-25

Servability

vLLM · SGLang

Servability

vLLM: supported · Gemma4ForConditionalGeneration · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · Gemma4ForConditionalGeneration · 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 gemma-4-E4B-it: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.

BF16 (2 bytes/param)19 GB · Single GPU (<24 GB)
FP8 (1 byte/param)9.6 GB · Single GPU (<24 GB)
INT4 (0.5 bytes/param)4.8 GB · Single GPU (<24 GB)

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

All variants

40 variants

Lineage

Fine-tuned from google/gemma-4-E4B (same organization) Family: google/gemma-4 1,266 derivatives on the Hub

Base model: google/gemma-4-E4B

finetune: 427quantized: 416adapter: 420merge: 3

Top variants (40 shown, ranked by downloads)

Relation
Quant
40 rows
Variant Relation Quant Params Band License Author Likes Downloads
farbodtavakkoli/OTel-LLM-E4B-IT finetune — — — apache-2.0 farbodtavakkoli 0 9.6M
lmstudio-community/gemma-4-E4B-it-MLX-4bit quantized MLX 8.00B M apache-2.0 lmstudio-community 35 6.9M
lmstudio-community/gemma-4-E4B-it-MLX-8bit quantized MLX 8.00B M apache-2.0 lmstudio-community 20 6.7M
lmstudio-community/gemma-4-E4B-it-MLX-6bit quantized MLX 8.00B M apache-2.0 lmstudio-community 17 6.6M
lmstudio-community/gemma-4-E4B-it-MLX-5bit quantized MLX 8.00B M apache-2.0 lmstudio-community 6 6.3M
unsloth/gemma-4-E4B-it-GGUF quantized GGUF — — apache-2.0 unsloth 664 6M
lmstudio-community/gemma-4-E4B-it-GGUF quantized GGUF — — apache-2.0 lmstudio-community 72 5.6M
unsloth/gemma-4-E4B-it finetune — 8.00B M apache-2.0 unsloth 27 3.9M
litert-community/gemma-4-E4B-it-litert-lm finetune — — — apache-2.0 litert-community 229 2.6M
unsloth/gemma-4-E4B-it-unsloth-bnb-4bit quantized -bnb 8.00B M apache-2.0 unsloth 23 2.3M
ggml-org/gemma-4-E4B-it-GGUF quantized GGUF — — apache-2.0 ggml-org 101 2.1M
cyankiwi/gemma-4-E4B-it-AWQ-INT4 quantized AWQ 7.94B M apache-2.0 cyankiwi 4 958.6k
OBLITERATUS/gemma-4-E4B-it-OBLITERATED quantized GGUF 8.00B M apache-2.0 OBLITERATUS 826 914.1k
llmfan46/gemma-4-E4B-it-uncensored-heretic finetune — 8.00B M apache-2.0 llmfan46 8 619.9k
bartowski/google_gemma-4-E4B-it-GGUF quantized GGUF — — apache-2.0 bartowski 85 508.5k
abhishekchohan/gemma-4-E4B-it-W4A16 quantized -W4A16 8.70B M — abhishekchohan 0 422.1k
leon-se/gemma-4-E4B-it-FP8-Dynamic quantized -FP8 8.67B M — leon-se 2 416.2k
ciocan/gemma-4-E4B-it-W4A16 quantized -W4A16 4.00B M apache-2.0 ciocan 3 412.3k
Chunity/gemma-4-E4B-it-AWQ-4bit quantized AWQ 4.81B M — Chunity 6 401.5k
mlx-community/gemma-4-e4b-it-4bit quantized -4bit 7.94B M gemma mlx-community 39 273.8k
prithivMLmods/gemma-4-E4B-it-NVFP4 quantized NVFP4 6.91B M apache-2.0 prithivMLmods 3 251.8k
Vishva007/gemma-4-E4B-it-W4A16-AutoRound-GPTQ quantized -W4A16 4.00B M gemma Vishva007 4 179.4k
bg-digitalservices/Gemma-4-E4B-it-NVFP4 quantized NVFP4 5.98B M apache-2.0 bg-digitalservices 1 163.7k
unsloth/gemma-4-E4B-it-NVFP4 quantized NVFP4 6.72B M apache-2.0 unsloth 14 160.3k
cosmicproc/gemma-4-E4B-it-NVFP4 quantized NVFP4 6.23B M apache-2.0 cosmicproc 14 153.9k
prithivMLmods/gemma-4-E4B-it-FP8 quantized -FP8 8.67B M apache-2.0 prithivMLmods 6 76.1k
mlx-community/gemma-4-e4b-it-8bit quantized -8bit 7.94B M gemma mlx-community 28 69.9k
coolthor/Gemma-4-E4B-it-NVFP4A16 quantized — 6.91B M apache-2.0 coolthor 1 64.3k
arsovskidev/Gemma-4-E4B-Claude-4.6-Opus-Reasoning-Distilled quantized GGUF 8.00B M apache-2.0 arsovskidev 22 59.7k
OpenVINO/gemma-4-E4B-it-int4-ov quantized int4 — — apache-2.0 OpenVINO 9 49.6k
google/gemma-4-E4B-it-qat-q4_0-unquantized finetune q4_0 7.94B M apache-2.0 google 43 41.8k
mlx-community/gemma-4-e4b-it-OptiQ-4bit quantized -4bit 7.46B M gemma mlx-community 29 40.9k
jgebbeken/gemma-4-coder-gguf quantized gguf — — apache-2.0 jgebbeken 22 40.9k
unsloth/gemma-4-E4B-it-UD-MLX-4bit quantized MLX 2.01B M apache-2.0 unsloth 46 35.7k
AtomicChat/gemma-4-E4B-it-GGUF quantized GGUF — — apache-2.0 AtomicChat 1 37.4k
MassivDash/Gemma-4-Rust-Coder quantized GGUF — — mit MassivDash 56 31.2k
Smoffyy/Gemma4-E4B-Instruct-Pure-GGUF quantized GGUF — — apache-2.0 Smoffyy 3 29.9k
TrevorJS/gemma-4-E4B-it-uncensored finetune — 8.00B M apache-2.0 TrevorJS 40 27.7k
google/gemma-4-E4B-it-qat-mobile-transformers quantized — 3.37B M apache-2.0 google 33 25.6k
Neural-ICE/Gemma-4-E4B-it-NVFP4 quantized NVFP4 5.98B M apache-2.0 Neural-ICE 0 24.6k

Repository changelog

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

Repository changelog 12 commits

DateCommitSHA
2026-07-20 Add response_template to tokenizer_config.json (#39) ee0ef60
2026-07-15 Update README (#42) fa62d88
2026-07-15 fix: chat template — null handling, reasoning preservation, turn-tag balance, input validation (#36) 6ea8ce9
2026-07-10 Add Technical Report a4c2d58
2026-06-03 Update README.md fee6332
2026-06-02 Update README.md cd0bc28
2026-05-18 fix(chat_template): Emit multimodal placeholders in tool response content-parts (#30) d6436b3
2026-05-07 Fix lineage (#31) 3555bdd
2026-04-28 fix(chat_template): update SI and tool call handling (#28) c53e9d3
2026-04-10 Update response schema (#16) 83df0a8
2026-04-10 fix: prefix-preserving tool calls + dialog compliance (#14) ab4d202
2026-04-02 Preparing for release! 292a7e2

Articles and coverage

37 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.

Official3

Gemma 4 QAT models: Optimizing compression for mobile and laptop efficiency
blog.google2026-06-05officialauthorityHN front pageabout the gemma 4 familyHN 406 pts · 130 comments714 words
Google released new Gemma 4 checkpoints optimized with Quantization-Aware Training to improve efficiency on edge devices and consumer GPUs. The E2B model achieves a 1GB memory footprint using a mobile-specialized schema that applies targeted 2-bit quantization to token generation layers. Developers can access GGUF formats ready for llama.cpp and MTP QAT checkpoints to preserve inference speedups while running locally.written, reviewed, cited
  • The mobile-specialized schema pre-calculates activation scaling settings during training to reduce workload on mobile chips.
  • Targeted 2-bit quantization compresses token generation parts while keeping core reasoning layers at higher precision.
  • Deploying only needed modalities allows the Gemma 4 E2B text-only model to require less than 1 GB of memory.
Accelerating Gemma 4: faster inference with multi-token prediction drafters
blog.google2026-05-05officialauthorityHN front pageabout the gemma 4 familyHN 687 pts · 330 comments796 words
Google released Multi-Token Prediction (MTP) drafters for the Gemma 4 family under the same open-source Apache 2.0 license as Gemma 4. These drafters deliver up to a 3x speedup for models like the 26B MoE and 31B Dense versions without quality degradation. Benchmarks on hardware such as the NVIDIA RTX PRO 6000 show that increasing batch sizes to 4 to 8 unlocks up to a ~2.2x speedup locally.written, reviewed, cited
  • The MTP drafter shares the target model's KV cache to avoid recalculating context the larger model has already processed.
  • For E2B and E4B edge models, an efficient clustering technique in the embedder accelerates generation where logit calculation is a bottleneck.
  • The 26B mixture-of-experts model presents unique routing challenges at a batch size of 1 on Apple Silicon.
Google releases Gemma 4 open models
deepmind.google2026-04-02officialauthorityHN front pageabout the gemma 4 familyHN 1,812 pts · 474 comments253 words

Analysis7

Show HN: Cactus Hybrid: We taught Gemma 4 to know when it's wrong
github.com2026-07-22authorityHN front pageabout the gemma 4 familyHN 191 pts · 44 comments714 words
Cactus Hybrid introduces on-device models that score every answer with a confidence between 0 and 1 to decide when to hand off to a bigger model. The Gemma 4 E2B Hybrid model matches Gemini 3.1 Flash-Lite performance by routing only 15–55% of queries to the cloud while running the rest locally. It is MIT-licensed and achieves a 0.770 AUROC on MMLU, with a probe trained on zero audio data yet reaching 0.79–0.88 AUROC on four audio benchmarks.written, reviewed, cited
  • Weights on the meta device crash the confidence read because probes score generations outside the module forward path.
  • The probe is a patch you compile into the llama.cpp engine rather than a standalone library component.
  • Transformers version 5.14 and above cause a segfault on this specific checkpoint.
Gemma 4 E2B inference in 700 lines of C
github.com2026-08-27authorityabout the gemma 4 familyHN 18 pts · 0 comments722 words
Fable 5 pushed Gemma 4 to 255 tok/s on WebGPU
xcancel.com2026-06-18about the gemma 4 familyHN 48 pts · 22 comments
Gemma 4 on iPhone
apps.apple.com2026-04-05HN front pageabout the gemma 4 familyHN 868 pts · 234 comments
Gemma 4 E4B running inside Obsidian via WebGPU and Karpathy-style LLM Wiki
github.com2026-09-09authorityabout the gemma 4 familyHN 4 pts · 1 comments
Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference
gizmoweek.com2026-04-15HN front pageabout the gemma 4 familyHN 303 pts · 187 comments364 words
Show HN: Determinstic LLM inference for lowest price Gemma 4, with Windows XP
tokendelivery.ai2026-09-12about the gemma 4 familyHN 9 pts · 1 comments
All coverage tracked for this model (37) — append-only, nothing is ever removed
TitleSourcePublishedFirst seenSignals
Gemma 4 Architecture: How 31B Beats 400B Modelskriraai.com—2026-10-07exa
Gemma 4 E4B IT Specs & Benchmarks | TheModelverse | The Modelversethemodelverse.in—2026-10-07exa
Show HN: Determinstic LLM inference for lowest price Gemma 4, with Windows XPtokendelivery.ai2026-09-122026-10-07hn
Gemma 4 E4B running inside Obsidian via WebGPU and Karpathy-style LLM Wikigithub.com2026-09-092026-10-07hn
Gemma 4 E2B inference in 700 lines of Cgithub.com2026-08-272026-10-07hn
What it takes to run Gemma 4 E4B on an iPhoneamoli.app2026-08-252026-10-07hn
Show HN: Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Macgithub.com2026-07-292026-10-07hn
Show HN: Cactus Hybrid: We taught Gemma 4 to know when it's wronggithub.com2026-07-222026-10-07hn
Running Gemma 4 26B at 5 tokens/sec on a 13-year-old Xeon with no GPUneomindlabs.com2026-07-152026-10-07hn
Show HN: Local privacy-first Microsoft Recall alternative with Gemma 4github.com2026-07-042026-10-07hn
Show HN: Running Gemma-4 26B at 124 tokens/SEC on a CPU, no GPUapeg.dev2026-06-302026-10-07hn
Gemma 4 on Cerebras - The Fastest Inference Is Now Multimodalcerebras.ai2026-06-302026-10-07hn
Google DeepMind’s Gemma 4: MoE, Efficiency Tricks, and...daily.dev2026-06-222026-10-07exa
Fable 5 pushed Gemma 4 to 255 tok/s on WebGPUxcancel.com2026-06-182026-10-07hn
Gemma 4 QAT models: Optimizing compression for mobile and laptop efficiencyblog.google official2026-06-052026-10-07hn
Google's new Gemma 4 12B model is designed to run on any laptop with 16GB of RAMarstechnica.com2026-06-032026-10-07hn
A Visual Guide to Gemma 4 12Bnewsletter.maartengrootendorst.com2026-06-032026-10-07hn
Gemma 4 12B: A unified, encoder-free multimodal modelblog.google official2026-06-032026-10-07hn
Show HN: I made a Gemma 4 Mac app that names screenshots with local AIsnapname.app2026-05-312026-10-07hn
Show HN: Local LLM code-generation with Gemma 4 e2B via JSON AST to Clojuregithub.com2026-05-192026-10-07hn
Accelerating Gemma 4: faster inference with multi-token prediction draftersblog.google official2026-05-052026-10-07hn
Show HN: Prompt-to-Excalidraw demo with Gemma 4 E2B in the browser (3.1GB)teamchong.github.io2026-04-192026-10-07hn
Understanding Gemma 4: A Deep Dive into Google's New Open-Weight Model Lineup - Dev Repositorydev-repository.com2026-04-162026-10-07exa
Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inferencegizmoweek.com2026-04-152026-10-07hn
Show HN: I benchmarked Gemma 4 E2B – the 2B model beat the 12B on multi-turnaiexplr.com2026-04-132026-10-07hn
I ran Gemma 4 as a local model in Codex CLIblog.danielvaughan.com2026-04-122026-10-07hn
Gemma 4 E2B vs E4B Benchmark: The Hidden Thinking Mode That Makes the Smaller Model 20× Slower — the KodeLabthekodelab.com2026-04-112026-10-07exa
A Visual Guide to Gemma 4newsletter.maartengrootendorst.com2026-04-092026-10-07hn
Gemma 4 Benchmarks: Ultimate Local AI Performance Guide 2026 - Gemma 4 Wikigemma4.wiki2026-04-092026-10-07exa
Show HN: Gemma 4 Multimodal Fine-Tuner for Apple Silicongithub.com2026-04-072026-10-07hn
Gemma 4 on iPhoneapps.apple.com2026-04-052026-10-07hn
Running Gemma 4 locally with LM Studio's new headless CLI and Claude Codeai.georgeliu.com2026-04-052026-10-07hn
April 2026 TLDR Setup for Ollama and Gemma 4 26B on a Mac minigist.github.com2026-04-032026-10-07hn
Gemma 4 Deep Dive — How Apache 2.0 Opens the Door to Sovereign AI | Pebblousblog.pebblous.ai2026-04-032026-10-07exa
Gemma 4 Deep Dive: Architecture, Benchmarks, and Impact — Swfte AIswfte.com2026-04-032026-10-07exa
Google releases Gemma 4 open modelsdeepmind.google official2026-04-022026-10-07hn
Gemma 4: Byte for byte, the most capable open modelsblog.google official2026-04-022026-10-07hn

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

  • MiniCPM-o-2_6 · 8.67B
    same size band (M), same task (any-to-any), permissive license
  • MiniCPM-o-4_5 · 9.37B
    same task (any-to-any), permissive license
  • Qwen2.5-Omni-7B · 10.7B
    same task (any-to-any), permissive license

Sources

3 sources

Sources

Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:32 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:45 UTC · Summary: 2026-10-07 15:00 UTC · Last verified 2026-10-25

Follow gemma-4-E4B-it

Watch this model

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