Llama-2-13b-chat-hf

meta-llama Llama 2 Community License out 3 years ago

13.0B parameters, text in, text out. 1 provider serves it.

Verified 2026-10-25 Rank #93 in LLMs · score 100.2 method v1 Sources 3 Downloads 30d 177.8k · #261 of 619 tracked

Can I run it?7.8GB at INT4Estimatefits a 16 GB cardmethodCommunity: Llama 2 13B Chat Hf by meta-llama — VRAM 26GB, 4K context | LLM Explorer →

Can I use it?Llama 2 Community LicenseAttribution; >700M MAU clause; acceptable-use policylicence file2026-10-24

Do people use it?177.8kdownloads, 30 days#261 of 619 we track · top 42%+3 likes, downloads −6% this week1,135 likesHugging Face2026-10-24

Worth reading

Picked by rule from 36 tracked pieces: the lab's own post, the most shared, a field report, the latest.

Most discussed · 955 HN pts

Fine-tune your own Llama 2 to replace GPT-3.5/4

Fine-tuning open-source Llama 2 models can replace GPT-3.5 and GPT-4 for specific tasks, with a fine-tuned Llama 7B matching GPT-4 labels 95% of the time on a test set.

news.ycombinator.com 2023-09-12 955 HN pts · 181 comments 10k words about the Llama 2 family family

Latest · 3 years ago

Run Llama 2 uncensored locally

The uncensored model provides a specific recipe for dangerously spicy mayo including ingredients like cayenne pepper and paprika.
The uncensored model describes the manufacturing process for Tylenol by combining paracetamol with aspirin, caffeine, and diphenhydramine.

ollama.ai 2023-08-02 722 HN pts · 212 comments 1.3k words about the Llama 2 family family

25 best of 36 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.

ArticleLlama 2ai.meta.com · 2023-07-18 · 2,268 HN pts · official
CodeLlama2.c: Inference llama 2 in one file of pure Cgithub.com · 2023-07-23 · 707 HN pts
CodeShow HN: Khoj – Chat offline with your second brain using Llama 2github.com · 2023-07-30 · 565 HN pts
CodeM2 Ultra can run 128 streams of Llama 2 7B in parallelgithub.com · 2023-10-11
ArticleGuide to running Llama 2 locallyreplicate.com · 2023-07-25 · 683 HN pts
ArticleAccessing Llama 2 from the command-line with the LLM-replicate pluginsimonwillison.net · 2023-07-18 · 200 HN pts
ArticleUnderstanding Llama 2 and the New Code Llama LLMsmagazine.sebastianraschka.com · 2023-08-30 · 170 HN pts
CodeLlama 2 Everywhere (L2E): Standalone, Binary Portable, Bootable Llama 2github.com · 2023-10-05 · 320 HN pts
CodeShow HN: LlamaGPT – Self-hosted, offline, private AI chatbot, powered by Llama 2github.com · 2023-08-16 · 225 HN pts
PaperLlama 2 Longarxiv.org · 2023-09-29
CodeLlama 2 on ONNX runs locallygithub.com · 2023-08-10
PaperLlamaF: An Efficient Llama2 Architecture Accelerator on Embedded FPGAsarxiv.org · 2024-09-27
ArticleLLaMA2 Chat 70B outperformed ChatGPTtatsu-lab.github.io · 2023-07-27
ArticleFast and Portable Llama2 Inference on the Heterogeneous Edgesecondstate.io · 2023-11-13
ArticleJetMoE: Reaching LLaMA2 performance with 0.1M dollarsresearch.myshell.ai · 2024-04-04
CodeShow HN: Llama2 Embeddings FastAPI Servergithub.com · 2023-08-15
ArticleFine-Tuning Llama-2: A Comprehensive Case Study for Tailoring Custom Modelsanyscale.com · 2023-08-11
ArticleA simple guide to fine-tuning Llama 2brev.dev · 2023-07-24
ArticleWhat's new in Llama 2 and how to run it locallyagi-sphere.com · 2023-08-06
DiscussionLeCun: Qualcomm working with Meta to run Llama-2 on mobile devicestwitter.com · 2023-07-23
ArticleLoRA Fine-Tuning Efficiently Undoes Safety Training from Llama 2-Chat 70Blesswrong.com · 2023-10-13
ArticleFast Llama 2 on CPUs with Sparse Fine-Tuning and DeepSparseneuralmagic.com · 2023-11-23
DiscussionVicuna v1.5 series, featuring 4K and 16K context, based on Llama 2twitter.com · 2023-08-03

Show all 23

Where to run it

Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.

#	Cheapest live row	In /1M	Out /1M	Context	Quant	Seen
1	Featherless	—	—	—	—	Oct 25

All 1 provider →

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.

Served by 1 API provider; 21 quantized builds for local use; a big base for builders (197 fine-tunes, adapters and merges).

As a service1Featherless
On own hardware21quantized builds
Built on19744 fine-tunes · 150 adapters · 3 merges
In apps100+Hugging Face Spaces

Top Spaces running it: huggingface-projects/llama-2-13b-chat ♥ 493 · h2oai/h2ogpt-chatbot ♥ 295 · baconnier/prompt-plus-plus ♥ 246 · allenai/WildBench ♥ 232

Will it fit?

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

Fits a 16 GB card at INT4

About 7.8 GB at INT4, 16 GB at FP8, 31 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.

10 dated articles and 0 repo commits so far

Teal is the weights (2023-07-13), amber triangles are 10 dated articles (each links to it).

What people built on it

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

218 derivatives on the Hub, most of them adapters

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

GGUF TheBloke/Llama-2-13B-chat-GGUF for llama.cpp, Ollama, LM Studio 474k downloads GPTQ TheBloke/Llama-2-13B-chat-GPTQ for vLLM, SGLang 1.7M downloads

All variants →

Reference

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

Use-case evidence

1 use case

What it is good at 1 use case

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 instructionsstructuraldecision-model estimate p=0.97 name: chat tag: conversational

Facts

17 fields

Listed in LLMs

Facts

Llama-2-13b-chat-hf has 13.0B parameters. (safetensors metadata, observed 2026-10-24)
License: Llama 2 Community License (friction: attribution). Attribution; >700M MAU clause; acceptable-use policy. (license file, observed 2026-10-24)
Created 2023-07-13, last modified 2024-04-17. (HF API, observed 2026-10-24)
Access is gated on Hugging Face (manual). (HF API, observed 2026-10-24)

Pipeline tag	text-generation
Library	transformers
Architectures	LlamaForCausalLM
Model type	llama
Params by dtype	F32: 2,560 · F16: 13,015,864,320
Size band	L
MoE	no
Gated	manual
Papers	arXiv:2307.09288
License	Llama 2 Community License (llama2)
Hub snapshot	2026-10-25

Servability

vLLM · SGLang

Servability

vLLM: supported · LlamaForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · LlamaForCausalLM · 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 Llama-2-13b-chat-hf: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.

BF16 (2 bytes/param)	31 GB · Single high-end GPU (<80 GB)
FP8 (1 byte/param)	16 GB · Single GPU (<24 GB)
INT4 (0.5 bytes/param)	7.8 GB · Single GPU (<24 GB)

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

All variants

40 variants

Lineage

Original model Family: meta-llama/Llama-2 218 derivatives on the Hub

quantized: 21finetune: 44adapter: 150merge: 3

Top variants (40 shown, ranked by downloads)

Variant	Relation	Quant	Params	Band	License	Author	Likes	Downloads
TheBloke/Llama-2-13B-chat-GPTQ	quantized	GPTQ	13.0B	L	llama2	TheBloke	362	1.7M
TheBloke/Llama-2-13B-chat-GGUF	quantized	GGUF	—	—	llama2	TheBloke	206	474k
TheBloke/Llama-2-13B-chat-AWQ	quantized	AWQ	13.0B	L	llama2	TheBloke	26	352.4k
TheBloke/Llama-2-13B-chat-GGML	finetune	—	—	—	other	TheBloke	692	30.8k
second-state/Llama-2-13B-Chat-GGUF	quantized	GGUF	—	—	llama2	second-state	3	21.4k
Nekochu/Luminia-13B-v3	adapter	GGUF	13.0B	L	apache-2.0	Nekochu	6	6.5k
mlc-ai/Llama-2-13b-chat-hf-q4f16_1-MLC	quantized	—	—	—	—	mlc-ai	1	4.3k
Liuzzyy/Llama-2-13b-chat-hf-Q4_K_M-GGUF	quantized	GGUF	—	—	llama2	Liuzzyy	0	3.1k
tensorblock/Llama-2-13b-chat-hf-GGUF	quantized	GGUF	—	—	llama2	tensorblock	0	2.1k
wentao-yuan/robopoint-v1-llama-2-13b	finetune	—	13.3B	L	apache-2.0	wentao-yuan	0	1.4k
FriendliAI/Llama-2-13b-chat-hf-fp8	finetune	-fp8	13.0B	L	llama2	FriendliAI	8	695
Thimira/sinhala-llama-2-13b-chat-hf	adapter	—	—	—	llama2	Thimira	0	829
NikolayKozloff/Llama-portuguese-13b-Luana-v0.2-GGUF	quantized	GGUF	—	—	apache-2.0	NikolayKozloff	1	685
kalpsnuti/llama-213-chat-gguf	quantized	gguf	—	—	llama2	kalpsnuti	0	729
rhaymison/Llama-portuguese-13b-Luana-v0.2	finetune	—	13.0B	L	apache-2.0	rhaymison	3	529
NikitaZagainov/notebook-segmentation-llama-2-13b-1ep	adapter	—	—	—	—	NikitaZagainov	0	629
wvnvwn/llama-2-13b-chat-hf-lr5e-5-resta-0.1	merge	—	13.0B	L	—	wvnvwn	0	619
BLACKBUN/llama-2-13b-pubmed-qa-211k-gguf_q8_0	quantized	gguf	—	—	llama2	BLACKBUN	2	421
AnirbanSaha/llama2-13b-tlink-sentence-only-autoregressive	finetune	—	13.0B	L	llama3.2	AnirbanSaha	0	510
wvnvwn/llama-2-13b-chat-hf-lr5e-5-resta-0.5	merge	—	13.0B	L	—	wvnvwn	0	491
wvnvwn/llama-2-13b-chat-hf-lr5e-5-resta-0.3	merge	—	13.0B	L	—	wvnvwn	0	485
zhentaoyu/Llama-2-13b-chat-hf-Q4_0-GGUF	quantized	Q4_0	—	—	llama2	zhentaoyu	0	464
nayohan/corningQA-llama2-13b-chat	finetune	—	—	—	apache-2.0	nayohan	1	338
leliuga/Llama-2-13b-chat-hf-bnb-4bit	quantized	-bnb	13.4B	L	apache-2.0	leliuga	0	372
vincentmin/llama-2-13b-reward-oasst1	adapter	—	—	—	—	vincentmin	0	368
irlab-udc/LLaMA2-13b-Stop-Hate	adapter	—	—	—	llama2	irlab-udc	1	252
BLACKBUN/llama-2-13b-pubmed-qa-211k	finetune	—	13.0B	L	mit	BLACKBUN	1	230
massh3dpotato/Llama-2-7B-chat-GGML	finetune	—	—	—	other	massh3dpotato	0	262
llmware/llama-2-13b-chat-ov	quantized	—	—	—	llama2	llmware	1	207
andrewrreed/Llama-2-13B-chat-GPTQ	quantized	GPTQ	13.0B	L	llama2	andrewrreed	0	253
massh3dpotato/Llama-2-13B-chat-GGML	finetune	—	—	—	other	massh3dpotato	0	242
jengjeeng/JengJeeng-R0-20250305	adapter	—	—	—	—	jengjeeng	1	156
mbkim/LifeTox_Moderator_13B	adapter	—	—	—	mit	mbkim	1	153
itsanurag/Llama-2-13b-Chat-4BitQuantized	quantized	-4Bit	13.4B	L	apache-2.0	itsanurag	1	147
NingLab/CASLIE-L	finetune	—	13.0B	L	cc-by-4.0	NingLab	0	173
jorgedelpozolerida/Llama-2-13b-chat-hf-Q8_0-GGUF	quantized	Q8_0	—	—	llama2	jorgedelpozolerida	0	172
boapro/Coder	quantized	GGUF	—	—	mit	boapro	1	113
jerife/llama2-13b-chat-hf-dbe-difficulty	adapter	—	—	—	—	jerife	0	148
AnirbanSaha/llama2-13b-tlink-full-autoregressive	finetune	—	13.0B	L	llama3.2	AnirbanSaha	0	144
HachiML/Llama-2-13b-chat-hf-ja-qlora	adapter	—	—	—	—	HachiML	0	142

All providers

1 provider · 1 endpoint

Where to run 1 providers · 1 endpoints

Hosted endpoints observed for Llama-2-13b-chat-hf, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting.

Provider	Model ref	Context	Quant	$/M in	$/M out	tok/s	TTFT	Tools	Status	Observed
Featherless hf	meta-llama/Llama-2-13b-chat-hf	—	—	—	—	—	—	—	live	2026-10-25

Articles and coverage

36 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.

Official1

Llama 2

ai.meta.com2023-07-18officialauthorityHN front pageabout the Llama 2 familyHN 2,268 pts · 820 comments42 words

Analysis8

Llama2.c: Inference llama 2 in one file of pure C

github.com2023-07-23authorityHN front pageabout the Llama 2 familyHN 707 pts · 165 comments4,774 words

The llama2.c repository offers a 700-line pure C file for training and inferring Llama 2 models, currently supporting fp32 inference for models up to 7B parameters. Work on model quantization is currently ongoing, but the project includes a 110M parameter model trained on the TinyStories dataset.written, reviewed, cited

Models larger than 7B are unusably slow because the repo performs float32 inference.
The 13B+ model variant fails due to integer flow errors in pointer arithmetic.

Show HN: Khoj – Chat offline with your second brain using Llama 2

github.com2023-07-30authorityHN front pageabout the Llama 2 familyHN 565 pts · 150 comments218 words

Khoj is an open-source, self-hostable AI platform that lets users get answers from the web or their docs. It enables building custom agents and scheduling automations while supporting online or local LLMs like gpt, claude, gemini, llama, qwen, and mistral. Users can get started for free and contribute to the community-sustained project.written, reviewed, cited

Run Llama 2 uncensored locally

ollama.ai2023-08-02HN front pageabout the Llama 2 familyHN 722 pts · 212 comments1,251 words

Ollama enables local execution of uncensored Llama 2 variants, such as the 7B model fine-tuned using Wizard-Vicuna conversation dataset and Nous Hermes Llama 2 13B fine-tuned on over 300,000 instructions. The 13B model stands out for its long responses, lower hallucination rate, and absence of OpenAI censorship mechanisms. Users can run these models via commands like ollama run llama2-uncensored to bypass standard safety refusals.written, reviewed, cited

The uncensored model provides a specific recipe for dangerously spicy mayo including ingredients like cayenne pepper and paprika.
The uncensored model describes the manufacturing process for Tylenol by combining paracetamol with aspirin, caffeine, and diphenhydramine.

Guide to running Llama 2 locally

replicate.com2023-07-25HN front pageabout the Llama 2 familyHN 683 pts · 170 comments603 words

Accessing Llama 2 from the command-line with the LLM-replicate plugin

simonwillison.net2023-07-18authorityHN front pageabout the Llama 2 familyHN 200 pts · 43 comments1,144 words

Understanding Llama 2 and the New Code Llama LLMs

magazine.sebastianraschka.com2023-08-30authorityHN front pageabout the Llama 2 familyHN 170 pts · 34 comments4,048 words

Llama 2 Everywhere (L2E): Standalone, Binary Portable, Bootable Llama 2

github.com2023-10-05authorityHN front pageabout the Llama 2 familyHN 320 pts · 55 comments4,153 words

Show HN: LlamaGPT – Self-hosted, offline, private AI chatbot, powered by Llama 2

github.com2023-08-16authorityHN front pageabout the Llama 2 familyHN 225 pts · 75 comments

Community1

Fine-tune your own Llama 2 to replace GPT-3.5/4

news.ycombinator.com2023-09-12HN front pageabout the Llama 2 familyHN 955 pts · 181 comments10,023 words

Fine-tuning open-source Llama 2 models can replace GPT-3.5 and GPT-4 for specific tasks, with a fine-tuned Llama 7B matching GPT-4 labels 95% of the time on a test set. While fine-tuning requires 1000+ examples and is less flexible than prompting, a fine-tuned Llama 7B is 50x cheaper than GPT-3.5 per token, costing just $19 to classify 2M recipes versus $23k for GPT-4.written, reviewed, cited

All coverage tracked for this model (36) — append-only, nothing is ever removed

Title	Source	Published	First seen	Signals
LLAMA 2 Paper and Model Deep Dive — The GenAI Guidebook	ravinkumar.com	—	2026-10-07	exa
LLaMA-2 from the Ground Up - by Cameron R. Wolfe, Ph.D.	cameronrwolfe.substack.com	—	2026-10-07	exa
meta-llama/Llama-2-13b-chat-hf Benchmark Scores & Evals	benchmarklist.com	—	2026-10-07	exa
Llama-2-13B Benchmarking: Performance Analysis - Truefoundry	truefoundry.com	—	2026-10-07	exa
Llama 2 13B Chat – 4k context, open weights \| LLM Reference	llmreference.com	—	2026-10-07	exa
Llama 2 13B \| Open Laboratory	openlaboratory.com	—	2026-10-07	exa
Llama 2 13B Chat Hf by meta-llama — VRAM 26GB, 4K context \| LLM Explorer	llm-explorer.com	—	2026-10-07	exa
LlamaF: An Efficient Llama2 Architecture Accelerator on Embedded FPGAs	arxiv.org	2024-09-27	2026-10-07	hn
JetMoE: Reaching LLaMA2 performance with 0.1M dollars	research.myshell.ai	2024-04-04	2026-10-07	hn
Fast Llama 2 on CPUs with Sparse Fine-Tuning and DeepSparse	neuralmagic.com	2023-11-23	2026-10-07	hn
Fast and Portable Llama2 Inference on the Heterogeneous Edge	secondstate.io	2023-11-13	2026-10-07	hn
LoRA Fine-Tuning Efficiently Undoes Safety Training from Llama 2-Chat 70B	lesswrong.com	2023-10-13	2026-10-07	hn
M2 Ultra can run 128 streams of Llama 2 7B in parallel	github.com	2023-10-11	2026-10-07	hn
Llama 2 Everywhere (L2E): Standalone, Binary Portable, Bootable Llama 2	github.com	2023-10-05	2026-10-07	hn
Llama 2 Long	arxiv.org	2023-09-29	2026-10-07	hn
Fine-tune your own Llama 2 to replace GPT-3.5/4	news.ycombinator.com	2023-09-12	2026-10-07	hn
Understanding Llama 2 and the New Code Llama LLMs	magazine.sebastianraschka.com	2023-08-30	2026-10-07	hn
Llama 2 is about as factually accurate as GPT-4 for summaries and is 30X cheaper	anyscale.com	2023-08-29	2026-10-07	hn
WebLLM: Llama2 in the Browser	webllm.mlc.ai	2023-08-29	2026-10-07	hn
Show HN: LlamaGPT – Self-hosted, offline, private AI chatbot, powered by Llama 2	github.com	2023-08-16	2026-10-07	hn
Show HN: Llama2 Embeddings FastAPI Server	github.com	2023-08-15	2026-10-07	hn
Fine-Tuning Llama-2: A Comprehensive Case Study for Tailoring Custom Models	anyscale.com	2023-08-11	2026-10-07	hn
Llama 2 on ONNX runs locally	github.com	2023-08-10	2026-10-07	hn
What's new in Llama 2 and how to run it locally	agi-sphere.com	2023-08-06	2026-10-07	hn
Vicuna v1.5 series, featuring 4K and 16K context, based on Llama 2	twitter.com	2023-08-03	2026-10-07	hn
Cookbook: Finetuning Llama 2 in your own cloud environment, privately	blog.skypilot.co	2023-08-02	2026-10-07	hn
Run Llama 2 uncensored locally	ollama.ai	2023-08-02	2026-10-07	hn
Show HN: Khoj – Chat offline with your second brain using Llama 2	github.com	2023-07-30	2026-10-07	hn
LLaMA2 Chat 70B outperformed ChatGPT	tatsu-lab.github.io	2023-07-27	2026-10-07	hn
Guide to running Llama 2 locally	replicate.com	2023-07-25	2026-10-07	hn
A simple guide to fine-tuning Llama 2	brev.dev	2023-07-24	2026-10-07	hn
Llama2.c: Inference llama 2 in one file of pure C	github.com	2023-07-23	2026-10-07	hn
LeCun: Qualcomm working with Meta to run Llama-2 on mobile devices	twitter.com	2023-07-23	2026-10-07	hn
Accessing Llama 2 from the command-line with the LLM-replicate plugin	simonwillison.net	2023-07-18	2026-10-07	hn
Llama 2	ai.meta.com official	2023-07-18	2026-10-07	hn
Llama 2 13B — benchmarks, pricing & hardware fit · Model Beats	modelbeats.com	2023-07-18	2026-10-07	exa

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).

Llama 2: Open Foundation and Fine-Tuned Chat Models this model

Hugo Touvron, Louis Martin, Kevin Stone +65 · 2023-07-18 · PDF · HF paper page (252 upvotes)

Similar models

3 models

Similar models

Qwen3-14B · 14.8B
same size band (L), same task (text-generation), more providers, permissive license
phi-4 · 14.7B
same size band (L), same task (text-generation), more providers, permissive license
DeepSeek-R1-Distill-Qwen-14B · 14.8B
same size band (L), same task (text-generation), permissive license

Sources

3 sources

Sources

Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:34 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:45 UTC · Summary: 2026-10-07 15:01 UTC · Last verified 2026-10-25

Follow Llama-2-13b-chat-hf

Watch this model

/feeds/models/meta-llama--Llama-2-13b-chat-hf.xml(RSS 2.0) · every change to Llama-2-13b-chat-hf
/feeds/families/meta-llama--Llama-2.xml(RSS 2.0) · every change across the Llama-2 family

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