# gemma-4-31B-it-assistant by Google: Open-Weight Model Source: https://savrn.com/models/gemma-4-31b-it-assistant Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve gemma-4-31B-it-assistant (470M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them. | Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits | | --- | --- | --- | --- | --- | --- | | 16-bit | 0.9 GB | 1.1 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 0.5 GB | 0.6 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 0.2 GB | 0.3 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the [SAVRN Index](https://savrn.com/ai-index/pricing/gpus), read Oct 24, 2026. [gemma-4-31B-it-assistant on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/gemma-4-31b-it-assistant/gpus) ## SAVRN's Notes on gemma-4-31B-it-assistant This 470M-parameter model is Google's assistant companion for Gemma 4 31B instruct. It is designed to run alongside the larger model inside a supporting serving stack, not to be used by itself. Its memory cost is about 1 GB. Apache 2.0 allows commercial use. If you run Gemma 4 31B in production, follow Google's card for the supported setup and measure the effect on your own request mix. SAVRN Research, 2026-09-18 ## Model Card By Google, published under apache-2.0, revision 627c5ec1458b. [Hugging Face](https://huggingface.co/collections/google/gemma-4) | [GitHub](https://github.com/google-gemma) | [MTP Documentation](https://ai.google.dev/gemma/docs/mtp/mtp) | [Technical Report](https://savrn.com/papers/gemma-4-technical-report) License: [Apache 2.0](https://ai.google.dev/gemma/docs/gemma_4_license) | Authors: [Google DeepMind](https://deepmind.google/models/gemma/) [!Note] This model card is for the Multi-Token Prediction (MTP) drafters for the Gemma 4 models. MTP is implemented by extending the base model with a smaller, faster draft model. When used in a Speculative Decoding pipeline, the draft model predicts several tokens ahead, which the target model then verifies in parallel. This results in significant decoding speedups (up to 3x) while guaranteeing the exact same quality as standard generation, making these checkpoints perfect for low-latency and on-device applications. [Read the full model card (3,614 words)](https://savrn.com/models/gemma-4-31b-it-assistant/card) ## Configuration Architecture Gemma4AssistantForCausalLM Context length (tokens) 262,144 Layers 4 Hidden size 1,024 Feed-forward size 8,192 Attention heads 32 Key/value heads 16 Head dimension 256 Vocabulary size 262,144 Sliding window (tokens) 1,024 Model type gemma4_assistant ## Identity and Version Repository google/gemma-4-31B-it-assistant Publisher Google Task Any to any Modality Multimodal Library transformers Parameters 470M parameters Languages Not stated by the source Revision 627c5ec1458b9086b841a91e0512fd31fd2fbbf1 First published 2026-04-23 Last updated 2026-07-15 ## Files and Weights 7 files, 971.2 MB in total. The weights are 1 file totalling 939.0 MB in safetensors. Weights1 file · 939.0 MB Configuration2 files · 2.6 KB Tokenizer2 files · 32.2 MB Documentation1 file · 29.8 KB Repository1 file · 1.6 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 939.0 MB | 9f80df6099fa | | config.json | Configuration | 2.3 KB | — | | generation_config.json | Configuration | 307 B | — | | README.md | Documentation | 29.8 KB | — | | .gitattributes | Repository | 1.6 KB | — | | tokenizer.json | Tokenizer | 32.2 MB | 75a6583c1a41 | | tokenizer_config.json | Tokenizer | 822 B | — | ## License and Download License apache-2.0 Access Open weights, no gate Download size 939.0 MB [Download from Google](https://www.kaggle.com/models/google/gemma-4) Released by Google through Kaggle. [Read the license](https://www.apache.org/licenses/LICENSE-2.0). ## Built From - Described by [arXiv:2607.02770](https://savrn.com/papers/gemma-4-technical-report) ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 939.0 MB | | 16-bit | 0.9 GB | | 8-bit | 0.5 GB | | 4-bit | 0.2 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About gemma-4-31B-it-assistant ### How much GPU memory does gemma-4-31B-it-assistant need? About 1.1 GB at 16-bit and 0.3 GB at 4-bit: the weights (470M parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run gemma-4-31B-it-assistant on? At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists. ### Can I use gemma-4-31B-it-assistant commercially? Yes. gemma-4-31B-it-assistant is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors. ### What is gemma-4-31B-it-assistant's context length? 262,144 tokens, from the maximum position embeddings in its published configuration. ## Similar Models Model · Any to any ### [gemma-4-26B-A4B-it-assistant](https://savrn.com/models/gemma-4-26b-a4b-it-assistant) [Google](https://savrn.com/model-publishers/google) Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages. Featuring both Dense and Mixture-of-Experts (MoE) architectures, Gemma 4 is well-suited for tasks like text generation, coding, and reasoning. The models are available in five distinct sizes: E2B, E4B, 12B, 26B A4B, and 31B. Their diverse sizes make them deployable in environments ranging from… Open weights apache-2.0 420M parameters 262,144 tokens transformers [View model](https://savrn.com/models/gemma-4-26b-a4b-it-assistant) Model · Any to any ### [gemma-4-E4B-it-assistant](https://savrn.com/models/gemma-4-e4b-it-assistant) [Google](https://savrn.com/model-publishers/google) Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages. Featuring both Dense and Mixture-of-Experts (MoE) architectures, Gemma 4 is well-suited for tasks like text generation, coding, and reasoning. The models are available in five distinct sizes: E2B, E4B, 12B, 26B A4B, and 31B. Their diverse sizes make them deployable in environments ranging from… Open weights apache-2.0 79M parameters 131,072 tokens transformers [View model](https://savrn.com/models/gemma-4-e4b-it-assistant) Model · Any to any ### [gemma-4-12B-it-qat-GGUF](https://savrn.com/models/gemma-4-12b-it-qat-gguf) [Unsloth AI](https://savrn.com/model-publishers/unsloth) This model ships a Multi-Token Prediction drafter at the repo root (mtp-gemma-4-12B-it.gguf, a near-lossless smart Q40). A recent llama.cpp auto-discovers it from -hf, so you do not pass --model-draft: The drafter shares the target's KV cache and does not change the output (the target verifies every drafted token). See the MTP/ folder for the other precisions and explicit usage. Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context… Open weights apache-2.0 transformers [View model](https://savrn.com/models/gemma-4-12b-it-qat-gguf) Model · Any to any ### [gemma-4-E4B-it-GGUF](https://savrn.com/models/gemma-4-e4b-it-gguf) [GGML Org](https://savrn.com/model-publishers/ggml-org) Run with https://llama.app - https://huggingface.co/google/gemma-4-E4B-it - https://huggingface.co/google/gemma-4-E4B-it-assistant - https://huggingface.co/google/gemma-4-E4B-it-qat-q40-unquantized-assistant - https://huggingface.co/google/gemma-4-E4B-it-qat-q40-unquantized - add info - add dflash Open weights apache-2.0 [View model](https://savrn.com/models/gemma-4-e4b-it-gguf) Model · Any to any ### [gemma-4-12B-it-qat-q4_0-gguf](https://savrn.com/models/gemma-4-12b-it-qat-q4-0-gguf) [Google](https://savrn.com/model-publishers/google) Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages. Featuring both Dense and Mixture-of-Experts (MoE) architectures, Gemma 4 is well-suited for tasks like text generation, coding, and reasoning. The models are available in five distinct sizes: E2B, E4B, 12B, 26B A4B, and 31B. Their diverse sizes make them deployable in environments ranging from… Open weights apache-2.0 transformers [View model](https://savrn.com/models/gemma-4-12b-it-qat-q4-0-gguf) Model · Any to any ### [gemma-4-E4B-it-qat-q4_0-gguf](https://savrn.com/models/gemma-4-e4b-it-qat-q4-0-gguf) [Google](https://savrn.com/model-publishers/google) Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages. Featuring both Dense and Mixture-of-Experts (MoE) architectures, Gemma 4 is well-suited for tasks like text generation, coding, and reasoning. The models are available in five distinct sizes: E2B, E4B, 12B, 26B A4B, and 31B. Their diverse sizes make them deployable in environments ranging from… Open weights apache-2.0 [View model](https://savrn.com/models/gemma-4-e4b-it-qat-q4-0-gguf) ## Google [All models and datasets](https://savrn.com/model-publishers/google) ## Versions - [627c5ec1458b](https://savrn.com/models/gemma-4-31b-it-assistant/versions/627c5ec1458b) · current 2026-09-18 ## Explore More - [All any to any models](https://savrn.com/models/tasks/any-to-any) - [All models under apache-2.0](https://savrn.com/models/licenses/apache-2-0) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-20. - [Google website](https://ai.google.dev/) - [Hugging Face record](https://huggingface.co/google/gemma-4-31B-it-assistant) - [How the hub is built](https://savrn.com/model-hub/methodology)