RunsOnMy ModelsGPUs & MacsBy VRAMBest GPUCan I run…?OllamaStart hereAPI pricesVRAM calculatorNew

Will this AI model run on my GPU?

VRAM needs for 298 open models on 65 GPUs and Macs, using real GGUF file sizes and each model's own config. Plus live prices for 470 API models, with every price change logged. New models appear within hours of release.

Popular GPUs and Macs

RTX 3060 12GB

12 GB · 173 models fit at 8K context

RTX 4060 Ti 16GB

16 GB · 200 models fit at 8K context

RTX 3090

24 GB · 209 models fit at 8K context

RTX 4090

24 GB · 209 models fit at 8K context

RTX 5090

32 GB · 218 models fit at 8K context

RTX 5070 Ti

16 GB · 200 models fit at 8K context

RX 7900 XTX

24 GB · 209 models fit at 8K context

Mac M4 Pro 48GB

48 GB · 218 models fit at 8K context

Mac M4 Max 128GB

128 GB · 238 models fit at 8K context

Mac M3 Ultra 512GB

512 GB · 292 models fit at 8K context

All 65 GPUs and Macs → · Apple Silicon guide

Models that fit your VRAM

8 GB VRAM

141 models fit at 4-bit, 8K context · 7.6 GB usable

12 GB VRAM

161 models fit at 4-bit, 8K context · 11.4 GB usable

16 GB VRAM

167 models fit at 4-bit, 8K context · 15.2 GB usable

24 GB VRAM

203 models fit at 4-bit, 8K context · 22.8 GB usable

32 GB VRAM

209 models fit at 4-bit, 8K context · 30.4 GB usable

All VRAM sizes, including 48 GB and 64 GB →

New this week

SenseNova Loop 9B

sensenova · 9.4B · 6.4 GB at 4-bit · out Nov 17, 2026

KORMo Web 11B

KORMo-VL · 11B · 8.3 GB at 4-bit · out Nov 14, 2026

SingVL S1 4B

inclusionAI · 4.5B · 3.6 GB at 4-bit · out Nov 11, 2026

Can you run this week's new models? → · Weekly Local LLM Hardware Index (charts + CSV) →

Popular questions

Can I run gpt-oss-120b on my PC?

Card-by-card, with CPU offload and RAM needed

Can I run Llama 70B?

Which GPUs and Macs hold it, and how fast

Can I run Qwen3?

Every Qwen3 size, cheapest GPU and smallest Mac

Best GPU for local LLMs by budget

$300 to $2,000, ranked by VRAM and bandwidth

Mac vs PC for local LLMs

Memory vs speed on popular models

Guides by tool

New here? How to run an LLM locally

Hardware, tools and your first model

Ollama models: size and VRAM

Every matched Ollama tag, filterable by your GPU

LM Studio: what fits your GPU

Quants explained with real file sizes

llama.cpp offload calculator

-ngl and --n-cpu-moe commands with RAM and speed

Best GPU for AI

Every consumer card ranked for local models

GPU vs GPU

RTX 3060 vs 4060, 4090 vs 5090 and more

Newest open models

ModelParamsQ4_K_M · 8KOn HF since
SenseNova Loop 9B sensenova9.4B6.4 GBNov 17, 2026
KORMo Web 11B KORMo-VL11B8.3 GBNov 14, 2026
SingVL S1 4B inclusionAI4.5B3.6 GBNov 11, 2026
Nex N2.6 nex-agi321B188 GBNov 10, 2026
MiniMax H3 Image PE MiniMaxAI27B16.8 GBNov 8, 2026
Arete 27B NousResearch27B17.4 GBNov 6, 2026
Inkling 2 Small thinkingmachines594B346 GBNov 4, 2026
LongCat 2.5 meituan-longcat1.78T1014 GBNov 4, 2026
Marin Mochi 535B A23B open-athena535B308 GBNov 3, 2026
Kimi K3.1 Pro moonshotai2.78T1584 GBNov 2, 2026

Everything new this month →

Newest API models

ModelInput / 1MOutput / 1MListed
OpenAI: GPT-6.2 Astra Pro$8.00$40.00Nov 17, 2026
OpenAI: GPT-6.2 Astra Pro (batch)$4.00$20.00Nov 17, 2026
OpenAI: GPT-6.2 Astra$8.00$40.00Nov 17, 2026
OpenAI: GPT-6.2 Astra (batch)$4.00$20.00Nov 17, 2026
OpenAI: GPT-6.2 Sol Pro$2.00$10.00Nov 17, 2026
OpenAI: GPT-6.2 Sol Pro (batch)$1.00$5.00Nov 17, 2026
OpenAI: GPT-6.2 Sol$2.00$10.00Nov 17, 2026
OpenAI: GPT-6.2 Sol (batch)$1.00$5.00Nov 17, 2026

All 470 API prices → · Price change log

Why trust these numbers

Every figure is computed, not guessed: weights from real GGUF file sizes, KV cache from each model's config.json (handles GQA, MLA, sliding-window and hybrid attention, and MoE active vs. total parameters), plus a published overhead. The formulas are on the methodology page, and automated checks hold back any page whose numbers fail a sanity test.