# Llama-3.2-1B by Meta Llama: Open-Weight Model Source: https://savrn.com/models/llama-3-2-1b Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve Llama-3.2-1B (1.2B 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 | 2.5 GB | 3.0 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 | 1.2 GB | 1.5 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 | 0.6 GB | 0.7 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. [Llama-3.2-1B on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/llama-3-2-1b/gpus) ## SAVRN's Notes on Llama-3.2-1B Three gigabytes at 16-bit, 1.5 GB at 8-bit, 0.7 GB at 4-bit: this 1.2B-parameter base model is small enough that the cheapest card on the Index, one MI300X at $1.85 an hour from Vultr, is overkill by two orders of magnitude. The realistic homes are a shared GPU already running other work, a workstation card, or the spare memory beside your main model. Note what the page does not list: a context length. Check it on the publisher's card before designing around a long prompt. Access is gated — files arrive after Meta approves a request — and the license is llama3.2, which this page does not summarize, so read the terms before shipping a product. The repository carries 4.9 GB across two weight files (safetensors and the original pth), and you need one. The lone evaluation on file, 11.95 on MMLU-Pro reported by a third party, tells you this is a base model, not a finished assistant; most operators should look one slot over at Llama-3.2-1B-Instruct. SAVRN Research ## Model Card The Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out). The Llama 3.2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks. They outperform many of the available open source and closed chat models on common industry benchmarks. Model Architecture: Llama 3.2 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for… Excerpt from the card by Meta Llama, licensed llama3.2. ## Identity and Version Repository meta-llama/Llama-3.2-1B Publisher Meta Llama Task Text generation Modality Text Library transformers Parameters 1.2B parameters Languages en, de, fr, it, pt, hi, es, th Revision 4e20de362430cd3b72f300e6b0f18e50e7166e08 First published 2024-09-18 Last updated 2024-10-24 ## Files and Weights 13 files, 5.0 GB in total. The weights are 2 files totalling 4.9 GB in pth, safetensors. Weights2 files · 4.9 GB Configuration4 files · 1.5 KB Tokenizer3 files · 11.3 MB Documentation3 files · 55.0 KB Repository1 file · 1.5 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 2.5 GB | — | | original/consolidated.00.pth | Weights | 2.5 GB | — | | config.json | Configuration | 843 B | — | | generation_config.json | Configuration | 185 B | — | | original/params.json | Configuration | 220 B | — | | special_tokens_map.json | Configuration | 301 B | — | | LICENSE.txt | Documentation | 7.7 KB | — | | README.md | Documentation | 41.2 KB | — | | USE_POLICY.md | Documentation | 6.0 KB | — | | .gitattributes | Repository | 1.5 KB | — | | original/tokenizer.model | Tokenizer | 2.2 MB | — | | tokenizer.json | Tokenizer | 9.1 MB | — | | tokenizer_config.json | Tokenizer | 50.5 KB | — | ## License and Download License llama3.2 Access Access requested at publisher Download size 4.9 GB [Download from Meta Llama](https://developer.meta.com/ai/llama-downloads/) Released by Meta Llama through Meta's Llama downloads. ## Built From - Described by [arXiv:2204.05149](https://savrn.com/papers/the-carbon-footprint-of-machine-learning-training-will-plateau-then-shrink) - Described by [arXiv:2405.16406](https://savrn.com/papers/spinquant-llm-quantization-with-learned-rotations) ## Evaluations Each result is shown as reported, with the conditions its reporter stated. None is a SAVRN measurement. A comparison lines two results up only when their configuration, unit and setup are all stated and identical. | Benchmark | Conditions | Result | Reported by | Revision | Date | | --- | --- | --- | --- | --- | --- | | TIGER-Lab/MMLU-Pro | Task mmlu_proMetric mmlu_proComparison conditions not established | 11.95 | [EvalEval](https://huggingface.co/datasets/evaleval/EEE_datastore/blob/192329fb7d6b15b7b0936a1a58ae862aa7e8ba24/flat/objects/d5/a3/d5a3c88a-3fee-40b3-a8c7-8c4cc65baf5f.json) Reported by a third party | Evaluated revision not stated | 2026-06-30 | ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 4.9 GB | | 16-bit | 2.5 GB | | 8-bit | 1.2 GB | | 4-bit | 0.6 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About Llama-3.2-1B ### How much GPU memory does Llama-3.2-1B need? About 3 GB at 16-bit and 0.7 GB at 4-bit: the weights (1.2B parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run Llama-3.2-1B 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. ### What license is Llama-3.2-1B released under? llama3.2, as its publisher declares it. Read the license text before commercial use. ## Similar Models Model · Text generation ### [Llama-3.2-1B-Instruct](https://savrn.com/models/llama-3-2-1b-instruct) [Meta Llama](https://savrn.com/model-publishers/meta-llama) The Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out). The Llama 3.2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks. They outperform many of the available open source and closed chat models on common industry benchmarks. Model Architecture: Llama 3.2 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for… Access requested at publisher llama3.2 1.2B parameters transformers [View model](https://savrn.com/models/llama-3-2-1b-instruct) Model · Text generation ### [Llama-3.2-1B-dolma3_mix](https://savrn.com/models/llama-3-2-1b-dolma3-mix) [CHEN EN LO](https://savrn.com/model-publishers/asciibase64) This revision publishes the final raw checkpoint of raw6b-normuon-adam2x-20261018: 6,000,082,944 non-padding input tokens, 28,619 optimizer updates. The model was randomly initialized (residual-half, seed42); no pretrained weights or previous training checkpoint initialized this run. Model architecture and The full reserved 50,000-document set has not been evaluated for this checkpoint. PPL is exp(sum held-out NLL / sum valid prediction tokens), not a mean of subset PPLs. Documents are evaluated independently, with stride8191 for long documents. The entire original document split was shuffled with seed42 before tokenization; the last50,000 documents were held out and excluded from all… Open weights llama3.2 1.2B parameters 131,072 tokens transformers [View model](https://savrn.com/models/llama-3-2-1b-dolma3-mix) Model · Text generation ### [Llama-3.2-1B-DelaySentinel](https://savrn.com/models/llama-3-2-1b-delaysentinel) [Yuchi Wang](https://savrn.com/model-publishers/yuchiwang02) Building and evaluating AI on logistics data. Built with Llama. This checkpoint is a full-parameter fine-tune of meta-llama/Llama-3.2-1B-Instruct, published with the data split, evaluation results, and behavioral tests used to inspect it. The project connects a working training-and-inference pipeline with a retrospective audit of what its score demonstrates. The checkpoint and a depth-2 decision tree both score 100% on the same historical 200-row split. A rule using two supplied fields reproduces every label in the 1,000-row source table. Prompt rewrites reveal additional response failures, including sensitivity to irrelevant text. These findings make the checkpoint useful for studying… Open weights llama3.2 1.2B parameters 131,072 tokens transformers [View model](https://savrn.com/models/llama-3-2-1b-delaysentinel) Model · Text generation ### [Llama-3.2-1B-Instruct-FT-20Minuten](https://savrn.com/models/llama-3-2-1b-instruct-ft-20minuten) [Yanggangu](https://savrn.com/model-publishers/yanggangu) A FT expert from Table 1 of SMAT: Simple and Efficient Merge-Aware Training (training seed 42). SMAT trains experts that retain task performance while improving model merging. Built with Llama. The included Llama 3.2 license and acceptable-use policy apply. Open weights llama3.2 1.2B parameters 131,072 tokens transformers [View model](https://savrn.com/models/llama-3-2-1b-instruct-ft-20minuten) Model · Text generation ### [Llama-3.2-1B-Instruct-FT-NumGLUE-ds](https://savrn.com/models/llama-3-2-1b-instruct-ft-numglue-ds) [Yanggangu](https://savrn.com/model-publishers/yanggangu) A FT expert from Table 1 of SMAT: Simple and Efficient Merge-Aware Training (training seed 42). SMAT trains experts that retain task performance while improving model merging. Built with Llama. The included Llama 3.2 license and acceptable-use policy apply. Open weights llama3.2 1.2B parameters 131,072 tokens transformers [View model](https://savrn.com/models/llama-3-2-1b-instruct-ft-numglue-ds) Model · Text generation ### [Llama-3.2-1B-Instruct-SMAT-20Minuten](https://savrn.com/models/llama-3-2-1b-instruct-smat-20minuten) [Yanggangu](https://savrn.com/model-publishers/yanggangu) A SMAT expert from Table 1 of SMAT: Simple and Efficient Merge-Aware Training (training seed 42). SMAT trains experts that retain task performance while improving model merging. Built with Llama. The included Llama 3.2 license and acceptable-use policy apply. Open weights llama3.2 1.2B parameters 131,072 tokens transformers [View model](https://savrn.com/models/llama-3-2-1b-instruct-smat-20minuten) ## Meta Llama [All models and datasets](https://savrn.com/model-publishers/meta-llama) ## Versions - [4e20de362430](https://savrn.com/models/llama-3-2-1b/versions/4e20de362430) · current 2026-09-18 ## Explore More - [All text generation models](https://savrn.com/models/tasks/text-generation) - [All models under llama3.2](https://savrn.com/models/licenses/llama3-2) - [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 2026-10-19. - [Meta Llama website](https://developer.meta.com/ai/) - [Hugging Face record](https://huggingface.co/meta-llama/Llama-3.2-1B) - [How the hub is built](https://savrn.com/model-hub/methodology)