Best Local LLMs & Local Models (2026)

Community-voted rankings for the best local models — open-source LLMs for coding, math, reasoning, and more

Get our weekly newsletter on pricing changes, new releases, and tools.

Provider	Model	Input $/M	Output $/M	LIVECODEBENCH	MATH_HARD	GPQA
QW Qwen	Qwen2.5 Coder 7B	$0.200	$0.200	-	-	-
M Meta-llama	Code Llama 7B Instruct	$0.200	$0.200	-	-	-
MI Mistral AI	Snorkel Mistral PairRM DPO	$0.200	$0.200	-	-	-
QW Qwen	Qwen3 8B	$0.050	$0.200	-	-	-
FW Fireworks	Marin 8B Instruct	$0.000	$0.000	-	-	-
LQ Liquid	LiquidAI/LFM2-8B-A1B	$0.010	$0.020	-	-	-
LQ Liquid	LiquidAI/LFM2-2.6B	$0.010	$0.020	-	-	-
G Google	CodeGemma 2B	$0.100	$0.100	-	-	-
G Google	CodeGemma 7B	$0.200	$0.200	-	-	-
NO Nousresearch	Chronos Hermes 13B v2	$0.200	$0.200	-	-	-
BY Bytedance	UI-TARS 7B	$0.100	$0.200	-	-	-
-	GLM 4.1V 9B Thinking	$0.000	$0.000	-	-	-
AP Alfredpros	CodeLLaMa 7B Instruct Solidity	$0.800	$1.200	-	-	-
CO Cohere	Command R7B (12-2024)	$0.037	$0.150	-	-	-
E Essential AI	Rnj 1 Instruct	$0.150	$0.150	-	-	-
QW Qwen	Qwen2 VL 7B Instruct	$0.200	$0.200	-	-	-
MI Mistral AI	Mixtral 8x7B Instruct (HF)	$0.500	$0.500	-	-	-
A Allenai	Molmo 2 4B	$0.200	$0.200	-	-	-
D Deepcogito	Cogito v1 Preview Llama 3B	$0.100	$0.100	-	-	-
EL Eleutherai	Pythia 12B	$0.200	$0.200	-	-	-
M Meta-llama	Code Llama 13B	$0.200	$0.200	-	-	-
M Meta-llama	Code Llama 13B Instruct	$0.200	$0.200	-	-	-
M Meta-llama	Code Llama 13B Python	$0.200	$0.200	-	-	-
M Meta-llama	Code Llama 7B	$0.200	$0.200	-	-	-
MI Mistral AI	Mistral 7B OpenOrca	$0.200	$0.200	-	-	-
QW Qwen	Qwen2 7B Instruct	$0.200	$0.200	-	-	-
QW Qwen	Qwen2 VL 2B Instruct	$0.100	$0.100	-	-	-
QW Qwen	Qwen3 4B	$0.200	$0.200	-	-	-
NO Nousresearch	OpenHermes 2.5 Mistral 7B	$0.170	$0.170	-	-	-
A Allenai	Molmo 2 8B	$0.200	$0.200	-	-	-
QW Qwen	CodeQwen 1.5 7B	$0.200	$0.200	-	-	-
QW Qwen	Qwen2.5 1.5B Instruct	$0.100	$0.100	-	-	-
QW Qwen	Qwen2.5 7B	$0.200	$0.200	-	-	-
QW Qwen	Qwen2.5 Coder 0.5B	$0.200	$0.200	-	-	-
QW Qwen	Qwen2.5 Coder 0.5B Instruct	$0.200	$0.200	-	-	-
QW Qwen	Qwen2.5 Coder 1.5B	$0.200	$0.200	-	-	-
QW Qwen	Qwen2.5 Coder 3B Instruct	$0.100	$0.100	-	-	-
QW Qwen	Qwen3 0.6B	$0.200	$0.200	-	-	-
QW Qwen	Qwen3 4B Instruct 2507	$0.200	$0.200	-	-	-
D Deepcogito	Cogito v1 Preview Llama 8B	$0.200	$0.200	-	-	-
DS Deepseek	DeepSeek Coder 1.3B Base	$0.100	$0.100	-	-	-
DS Deepseek	DeepSeek Coder 7B Base	$0.200	$0.200	-	-	-
NO Nousresearch	Toppy M 7B	$0.200	$0.200	-	-	-
NO Nousresearch	Zephyr 7B Beta	$0.200	$0.200	-	-	-
AR Arcee AI	Trinity Mini	$0.045	$0.150	-	-	-
A Allenai	Olmo 3 7B Instruct	$0.100	$0.200	-	-	-
A Allenai	Olmo 3 7B Think	$0.120	$0.200	-	-	-
DS Deepseek	DeepSeek Coder 7B Base v1.5	$0.200	$0.200	-	-	-
DS Deepseek	DeepSeek Coder 7B Instruct v1.5	$0.200	$0.200	-	-	-
DS Deepseek	R1 Distill Qwen 1.5B	$0.200	$0.200	-	-	-
G Google	Gemma 2B	$0.100	$0.100	-	-	-
G Google	Gemma 7B	$0.200	$0.200	-	-	-
M Meta-llama	Llama 2 13B	$0.200	$0.200	-	-	-
M Meta-llama	Llama 2 13B Chat	$0.200	$0.200	-	-	-
NO Nousresearch	Hermes 2 Pro Mistral 7B	$0.200	$0.200	-	-	-
OG Opengvlab	InternVL3 8B	$0.200	$0.200	-	-	-
QW Qwen	Qwen2.5 Coder 1.5B Instruct	$0.200	$0.200	-	-	-
QW Qwen	Qwen2.5 Coder 3B	$0.100	$0.100	-	-	-
QW Qwen	Qwen2.5 VL 3B Instruct	$0.100	$0.100	-	-	-
QW Qwen	Qwen3 1.7B	$0.200	$0.200	-	-	-
MI Mistral AI	Ministral 3 8B 2512	$0.150	$0.150	-	-	-
MI Mistral AI	Ministral 3 3B 2512	$0.100	$0.100	-	-	-
NV Nvidia	Nemotron Nano 12B 2 VL	$0.000	$0.000	-	-	-
NV Nvidia	Nemotron Nano 9B V2	$0.000	$0.000	-	-	-
MI Mistral AI	Mistral 7B v0.2	$0.200	$0.200	-	-	-
M Meta-llama	Llama 2 7B Chat	$0.200	$0.200	-	-	-
M Meta-llama	Llama 3.2 1B	$0.100	$0.100	-	-	-
M Meta-llama	Llama 3.2 3B	$0.100	$0.100	-	-	-
M Meta-llama	Llama 3 8B	$0.050	$0.080	-	-	-
MI Mistral AI	Mistral 7B	$0.200	$0.200	-	-	-
NO Nousresearch	Nous Hermes 2 Mixtral 8x7B DPO	$0.500	$0.500	-	-	-
NO Nousresearch	Nous Hermes Llama 2 13B	$0.170	$0.170	-	-	-
NO Nousresearch	Nous Hermes Llama 2 7B	$0.200	$0.200	-	-	-
NO Nousresearch	OpenHermes 2 Mistral 7B	$0.200	$0.200	-	-	-
NV Nvidia	Nemotron Nano 12B V2	$0.200	$0.200	-	-	-
Z Z-ai	GLM 4.5V	$0.600	$1.800	-	-	-
A Allenai	OLMo 7B Instruct	$0.200	$0.200	-	-	-
A Austism	Austism/chronos-hermes-13b	$0.300	$0.300	-	-	-
DS Deepseek	DeepSeek R1 0528 Qwen3 8B	$0.200	$0.200	-	-	-
G Google	Gemma 3n 4B	$0.020	$0.040	-	-	-
M Meta-llama	Llama 2 7B	$0.200	$0.200	-	-	-
CG Cognitive Computations	Dolphin 2.6 Mixtral 8x7B	$0.500	$0.500	-	-	-
G Google	Gemma 7B Instruct	$0.070	$0.070	-	-	-
M Meta-llama	Llama 3 8B Instruct (HF)	$0.200	$0.200	-	-	-
MI Mistral AI	Mixtral 8x7B	$0.240	$0.240	-	-	-
NO Nousresearch	OpenChat 3.5 0106	$0.200	$0.200	-	-	-
QW Qwen	Qwen3.5 9B	$0.050	$0.150	-	-	-
U Upstage	SOLAR 10.7B Instruct v1	$0.300	$0.300	-	-	-
QW Qwen	Qwen2.5 Coder 7B Instruct	$0.030	$0.090	-	-	-
EL Eleutherai	Llemma 7b	$0.800	$1.200	-	-	-
G Google	Gemma 3 4B	$0.040	$0.080	-	-	-
G Google	Gemma 3 12B	$0.040	$0.130	-	-	-
MS Microsoft	Phi 4 Multimodal Instruct	$0.000	$0.000	-	-	-
TD Thedrummer	UnslopNemo 12B	$0.400	$0.400	-	-	-
MI Mistral AI	Ministral 8B	$0.100	$0.100	-	-	-
MI Mistral AI	Ministral 3B	$0.040	$0.040	-	-	-
G Google	Gemma 2B	$0.100	$0.100	-	-	-
IB Ibm	ALLaM 7B Instruct Preview	$0.000	$0.000	-	-	-
L Lmsys	lmsys/vicuna-13b-v1.5	$0.300	$0.300	-	-	-
QW Qwen	Qwen2.5 7B Instruct	$0.040	$0.100	-	-	-
NO Nousresearch	OpenChat 7B	$0.060	$0.060	-	-	-
TD Thedrummer	Rocinante 12B	$0.170	$0.430	-	-	-
M Meta-llama	Llama 3.2 3B Instruct	$0.030	$0.050	-	-	-
M Meta-llama	Llama 3.2 1B Instruct	$0.020	$0.020	-	-	-
M Meta-llama	Llama 3.2 11B Vision Instruct	$0.049	$0.049	-	-	-
QW Qwen	Qwen3 VL 8B Instruct	$0.080	$0.200	-	-	-
QW Qwen	Qwen1.5 0.5B	$0.100	$0.100	-	-	-
QW Qwen	Qwen1.5 0.5B Chat	$0.100	$0.100	-	-	-
NS Neversleep	Lumimaid v0.2 8B	$0.000	$0.000	-	-	-
MI Mistral AI	Pixtral 12B	$0.150	$0.150	-	-	-
QW Qwen	Qwen2.5-VL 7B Instruct	$0.200	$0.200	-	-	-
MS Microsoft	Phi-3.5 Mini 128K Instruct	$0.000	$0.000	-	-	-
S1 Sao10k	Llama 3 8B Lunaris	$0.040	$0.050	-	-	-
M Meta-llama	Llama 3.1 8B Instruct	$0.020	$0.050	-	-	-
MI Mistral AI	Mistral Nemo	$0.020	$0.040	-	-	-
NO Nousresearch	Hermes 2 Pro - Llama-3 8B	$0.140	$0.140	-	-	-
G Google	Gemma 2 9B	$0.030	$0.080	-	-	-
MI Mistral AI	Mistral 7B Instruct	$0.059	$0.059	-	-	-
MI Mistral AI	Mistral 7B Instruct v0.3	$0.140	$0.200	-	-	-
MS Microsoft	Phi-3 Mini 128K Instruct	$0.000	$0.000	-	-	-
MI Mistral AI	Mistral Tiny	$0.140	$0.420	-	-	-
M Meta-llama	Llama 3 8B Instruct	$0.030	$0.040	-	-	-
MI Mistral AI	Mistral 7B Instruct v0.2	$0.140	$0.200	-	-	-
MI Mistral AI	Mixtral 8x7B Instruct	$0.140	$0.420	-	-	-
U9 Undi95	ReMM SLERP 13B	$0.300	$0.300	-	-	-
GR Gryphe	MythoMax 13B	$0.060	$0.060	-	-	-
G Google	Gemma 3 1B	$0.100	$0.100	-	-	-
O Open Orca	Open-Orca/Mistral-7B-OpenOrca	$0.200	$0.200	-	-	-
QW Qwen	Qwen2.5 0.5B Instruct	$0.200	$0.200	-	-	-
NO Nousresearch	Nous Capybara 7B v1.9	$0.200	$0.200	-	-	-

Vote for open-source models that work well (or don't) for local use.

Pricing from OpenRouter.

Running LLMs Locally

Open-source models can run on your own hardware using tools like Ollama, llama.cpp, or vLLM. This gives you full privacy, zero API costs, and offline capability.

VRAM Requirements

7B models need ~6GB VRAM (4-bit quantized). 13B models need ~10GB. 70B models need ~40GB or multi-GPU.

Recommended: Ollama

Install Ollama, run ollama pull qwen3:8b, then point your tool at http://localhost:11434/v1.

Alternative: llama.cpp / vLLM

For more control over quantization and batched inference, these give you a full OpenAI-compatible API with fine-grained settings.

Compare all local LLM runners →

About This Leaderboard

This leaderboard ranks open-source, locally-runnable AI models by community votes from developers. Models are filtered to open-weight models with 13B parameters or fewer — small enough to run on a single consumer GPU.

Use the tabs above to see which local models are best for specific tasks like coding, math, or general reasoning. Benchmark scores from Artificial Analysis are shown alongside votes to help you compare real-world experience with synthetic performance.

Best Local LLMs & Local Models (2026)

Community-voted rankings for the best local models — open-source LLMs for coding, math, reasoning, and more

Running LLMs Locally

About This Leaderboard

Frequently Asked Questions

Tools

Directories

Models & Pricing

Endpoints

Rankings

News