OpenELM-1_1B-Instruct

apple apple-sample-code-license out 3 years ago

1.08B parameters, text in, text out.

Verified 2026-10-25 Rank #636 in LLMs · score 73.3 method v1 Sources 3 Downloads 30d 5.4k · #536 of 619 tracked

Can I run it?0.6GB at INT4Estimatefits a 16 GB cardmethod

Can I use it?apple-sample-code-licenseRead the terms firstlicence file2026-10-24

Do people use it?5.4kdownloads, 30 days#536 of 619 we track · below the median+0 likes, downloads −98% this week78 likesHugging Face2026-10-24

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: 9 quantized builds and 1 Spaces.

As a service0no API provider listed
On own hardware9quantized builds
Built on43 fine-tunes · 1 adapters · 0 merges
In apps1Hugging Face Spaces

Will it fit?

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

Fits a 16 GB card at INT4

About 0.6 GB at INT4, 1.3 GB at FP8, 2.6 GB at BF16: weights only, so leave room for the KV cache.

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

The story so far

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

6 repo commits since the weights landed

Teal is the weights (2024-04-12), grey ticks are repo commits.

What people built on it

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

13 derivatives on the Hub, most of them quantized

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

GGUF mradermacher/OpenELM-1_1B-Instruct-GGUF for llama.cpp, Ollama, LM Studio 11.1k downloads MLX uirev/MLX_OpenELM_1.1b_UiRev_finetuned_V3 for Apple silicon 216 downloads

All variants →

Reference

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

Benchmarks

23 rows

Benchmarks 23 rows

measured third-party leaderboard rowauthor-reported number from the model card or eval files (23)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.

Benchmark	Value	Setting / notes	Reported by	Date
author-reported ARC Challenge · Chat and follow instructions	37.97	Zero-Shot	model author	2026-10-07
author-reported ARC Challenge · Chat and follow instructions	41.55	LLM360	model author	2026-10-07
author-reported ARC Challenge · Chat and follow instructions	41.55	OpenLLM Leaderboard	model author	2026-10-07
author-reported Average	65.5	Zero-Shot	model author	2026-10-07
author-reported Average	49.94	LLM360	model author	2026-10-07
author-reported Average	54.4	OpenLLM Leaderboard	model author	2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions	52.23	Zero-Shot	model author	2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions	70	Zero-Shot	model author	2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions	75.03	Zero-Shot	model author	2026-10-07
author-reported Commonsense QA sets · Chat and follow instructions	75.03	OpenLLM Leaderboard	model author	2026-10-07
author-reported CrowS-Pairs	71.02	OpenLLM Leaderboard	model author	2026-10-07
author-reported HellaSwag · Chat and follow instructions	71.2	Zero-Shot	model author	2026-10-07
author-reported HellaSwag · Chat and follow instructions	71.83	LLM360	model author	2026-10-07
author-reported HellaSwag · Chat and follow instructions	71.83	OpenLLM Leaderboard	model author	2026-10-07
author-reported MMLU · Reason through hard problems	25.65	LLM360	model author	2026-10-07
author-reported MMLU · Reason through hard problems	25.65	OpenLLM Leaderboard	model author	2026-10-07
author-reported RACE	39.43	OpenLLM Leaderboard	model author	2026-10-07
author-reported SciQ	89.3	Zero-Shot	model author	2026-10-07
author-reported TruthfulQA · Chat and follow instructions	45.95	LLM360	model author	2026-10-07
author-reported TruthfulQA · Chat and follow instructions	45.95	OpenLLM Leaderboard	model author	2026-10-07
author-reported Winogrande · Chat and follow instructions	62.75	Zero-Shot	model author	2026-10-07
author-reported Winogrande · Chat and follow instructions	64.72	LLM360	model author	2026-10-07
author-reported Winogrande · Chat and follow instructions	64.72	OpenLLM Leaderboard	model author	2026-10-07

Use-case evidence

3 use cases

What it is good at 3 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.

Chat and follow instructionsauthor-reportedAuthor reports ARC-c 41.55 Author reports PIQA 75.03 Author reports HellaSwag 71.83 +4 more
Reason through hard problemsauthor-reportedAuthor reports MMLU 25.65
Run on a laptop or phonestructuraldecision-model estimate p=1.00 name: 1B fits: edge

Facts

17 fields

Listed in LLMs

Facts

OpenELM-1_1B-Instruct has 1.08B parameters. (safetensors metadata, observed 2026-10-24)
License: apple-sample-code-license (friction: review). (license file, observed 2026-10-24)
Created 2024-04-12, last modified 2025-02-28. (HF API, observed 2026-10-24)

Pipeline tag	text-generation
Library	transformers
Architectures	OpenELMForCausalLM
Model type	openelm
Dtype	bfloat16
Params by dtype	BF16: 1,079,891,456
Size band	S
MoE	no
Papers	arXiv:2404.14619
License	apple-sample-code-license (apple-amlr)
Hub snapshot	2026-10-25

Servability

vLLM · SGLang

Servability

vLLM: not in registry · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: not in registry · 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 OpenELM-1_1B-Instruct: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.

BF16 (2 bytes/param)	2.6 GB · Edge (<4 GB)
FP8 (1 byte/param)	1.3 GB · Edge (<4 GB)
INT4 (0.5 bytes/param)	0.6 GB · Edge (<4 GB)

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

All variants

13 variants

Lineage

Original model Family: apple/OpenELM-1 13 derivatives on the Hub

quantized: 9finetune: 3adapter: 1

Top variants (13 shown, ranked by downloads)

Variant	Relation	Quant	Params	Band	License	Author	Downloads
mradermacher/OpenELM-1_1B-Instruct-GGUF	quantized	GGUF	—	—	apple-amlr	mradermacher	11.1k
uirev/MLX_OpenELM_1.1b_UiRev_finetuned_V3	finetune	—	1.08B	S	other	uirev	216
DaveLeongSG/OpenELM-1_1B-Instruct-Q4_K_M-GGUF	quantized	GGUF	—	—	other	DaveLeongSG	187
marroyo777/OpenELM-1_1B-Instruct-Q4_K_M-GGUF	quantized	GGUF	—	—	other	marroyo777	185
fernandoruiz/OpenELM-1_1B-Instruct-Q4_K_M-GGUF	quantized	GGUF	—	—	other	fernandoruiz	183
ysn-rfd/OpenELM-1_1B-Instruct-Q8_0-GGUF	quantized	Q8_0	—	—	other	ysn-rfd	178
reach-vb/OpenELM-1_1B-Instruct-Q8_0-GGUF	quantized	Q8_0	—	—	other	reach-vb	169
fernandoruiz/OpenELM-1_1B-Instruct-Q4_K_S-GGUF	quantized	GGUF	—	—	other	fernandoruiz	156
uirev/MLX_OpenELM_1.1b_UiRev_finetuned_V2	finetune	—	1.08B	S	other	uirev	154
1bit-MONSTER/OpenELM-1.1B-Instruct-GGUF	quantized	GGUF	—	—	other	1bit-MONSTER	115
Grogros/OpenELM-1_1B-Instruct-distillation-alpaca-AlpacaPoison-tulu	finetune	—	1.08B	S	apple-amlr	Grogros	113
straino/OpenELM-1_1B-Instruct-Q4_K_M-GGUF	quantized	GGUF	—	—	apple-amlr	straino	113
ApdoElepe/openelm-safety-lora	adapter	—	—	—	apache-2.0	ApdoElepe	75

Repository changelog

6 commits

Repository changelog 6 commits

Date	Commit	SHA
2025-02-28	AMLR license (#4)	effd796
2025-02-24	Update LICENSE	a1d8525
2024-07-18	update OpenELM-1_1B-Instruct	f2baa5b
2024-05-02	update OpenELM-1_1B	24982bb
2024-04-24	add OpenELM-1_1B	654de97
2024-04-23	add OpenELM-1_1B-Instruct	b418636

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).

OpenELM: An Efficient Language Model Family with Open Training and Inference Framework this model

Sachin Mehta, Mohammad Hossein Sekhavat, Qingqing Cao +8 · 2024-04-22 · PDF · HF paper page (126 upvotes)

Similar models

3 models

Similar models

Llama-3.2-1B-Instruct · 1.24B
same size band (S), same task (text-generation), more providers
gemma-3-1b-it · 1000M
same size band (S), same task (text-generation), more providers
TinyLlama-1.1B-Chat-v1.0 · 1.10B
same size band (S), same task (text-generation), more providers, permissive license

Sources

3 sources

Sources

Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:38 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:09 UTC · Last verified 2026-10-25

Follow OpenELM-1_1B-Instruct

Watch this model

/feeds/models/apple--OpenELM-1_1B-Instruct.xml(RSS 2.0) · every change to OpenELM-1_1B-Instruct
/feeds/families/apple--OpenELM-1.xml(RSS 2.0) · every change across the OpenELM-1 family

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