What can you run on a 16GB graphics card?

With 16GB VRAM you can run 46 of the 101 open-source models we track — up to about 22B parameters at Q4 quantization. The most popular that fits is Qwen3.5-9B, needing 6.8 GB and generating around ~82 tok/s on typical hardware of this size.

Typical machines: RTX 4080 · RTX 5080 · RTX 4060 Ti 16GB · RX 7800 XT · Arc A770. After the operating system takes its share, about 16 GB is free for a model.

Having the memory is not the same as having the speed

16GB VRAM decides whether a model loads. How fast it answers is decided by memory bandwidth — how quickly your machine can read the model, which it must do once for every word it writes. Machines with 16GB VRAM span 288–960 GB/s, so the fastest is about 3.3× quicker than the slowest with identical capacity. The table quotes a RTX 4080; pick yours below to see the difference. Why bandwidth and not the processor →

288 GB/sGeForce RTX 4060 Ti (16GB)· NVIDIA published specification: 288 GB/s GDDR6, 128-bit
560 GB/sIntel Arc A770 (16GB)· Intel published specification: 560 GB/s GDDR6, 256-bit
624 GB/sRadeon RX 7800 XT (16GB)· AMD published specification: 624 GB/s GDDR6, 256-bit
717 GB/sGeForce RTX 4080 (16GB)· NVIDIA published specification: 716.8 GB/s GDDR6X, 256-bit
960 GB/sGeForce RTX 5080 (16GB)· NVIDIA published specification: 960 GB/s GDDR7, 256-bit

Models that run on 16GB VRAM in 2026

ModelSizeMemoryEst. speedDownloadsFeels like
granite-embedding-97m-multilingual-r2Ibm97M655 MBRuns well~4419 tok/s237,787Fast
granite-embedding-311m-multilingual-r2Ibm312M748 MBRuns well~2477 tok/s47,167Fast
LFM2.5-Encoder-230MLiquid Ai230M793 MBRuns well~3361 tok/s13,450Fast
LFM2.5-230MLiquid Ai230M808 MBRuns well~3038 tok/s54,986Fast
ERNIE-4.5-0.3B-PTBaidu AI361M853 MBRuns well~1937 tok/s21,323Fast
nemotron-3.5-asr-streaming-0.6bNVIDIA638M1.1 GBRuns well~940 tok/s1,052,774Fast
Qwen3.5-0.8BAlibaba / Qwen873M1.3 GBRuns well~875 tok/s2,990,867Fast
LFM2.5-1.2B-InstructLiquid Ai1.2B1.4 GBRuns well~638 tok/s555,431Fast
LFM2.5-VL-1.6BLiquid Ai1.6B1.4 GBRuns well~638 tok/s78,655Fast
Nemotron-3-Embed-1B-BF16NVIDIA1.1B1.5 GBRuns well~677 tok/s467,453Fast
cohere-transcribe-arabic-07-2026Cohere2.1B1.9 GBRuns well~374 tok/s51,389Fast
Hy-MT2-1.8BTencent2B1.9 GBRuns well~411 tok/s49,645Fast
Qwen3.5-2BAlibaba / Qwen2.3B2.0 GBRuns well~364 tok/s2,468,876Fast
granite-speech-4.1-2bIbm2.3B2.4 GBRuns well~299 tok/s402,320Fast
LFM2.5-2.6BLiquid Ai2.7B2.5 GBRuns well~278 tok/s77,973Fast
LFM2.5-Audio-1.5BLiquid Ai1.5B2.6 GBRuns well~256 tok/s1,025Fast
Unlimited-OCRBaidu AI3.3B2.7 GBRuns well~239 tok/s2,836,694Fast
LocateAnything-3BNVIDIA3.8B2.8 GBRuns well~221 tok/s463,853Fast
granite-4.1-3bIbm3.4B3.0 GBRuns well~222 tok/s300,243Fast
Shieldstral-1.0-3BMistral AI3.8B3.1 GBRuns well~217 tok/s2,480Fast
Ministral-3-3B-Instruct-2512Mistral AI3.8B3.1 GBRuns well~217 tok/s544,419Fast
Nemotron-Labs-Diffusion-3B-BaseNVIDIA3.8B3.3 GBRuns well~201 tok/s35,110Fast
Cosmos3-EdgeNVIDIA3.9B3.3 GBRuns well~200 tok/s53,138Fast
Nemotron-3.5-Content-SafetyNVIDIA4.3B3.6 GBRuns well~187 tok/s176,924Fast
gemma-4-E2B-itGoogle DeepMind5.1B3.8 GBRuns well~150 tok/s4,002,947Fast
Qwen3.5-4BAlibaba / Qwen4.7B3.8 GBRuns well~170 tok/s6,199,203Fast
Fara1.5-4BMicrosoft4.5B4.0 GBRuns well~162 tok/s4,882Fast
NVIDIA-Nemotron-3-Nano-4B-BF16NVIDIA4B4.1 GBRuns well~164 tok/s712,701Fast
Fara-7BMicrosoft8.3B5.5 GBRuns well~100 tok/s2,192Fast
Hy-MT2-7BTencent8B5.7 GBRuns well~101 tok/s26,629Fast
LFM2.5-8B-A1BLiquid Ai · MoE8.5B5.9 GBRuns well~765 tok/s1B active170,464Fast
Nemotron-3-Embed-8B-BF16NVIDIA8B5.9 GBRuns well~97 tok/s142,381Fast
gemma-4-E4B-itGoogle DeepMind8B6.2 GBRuns well~87 tok/s5,295,850Fast
Nemotron-Labs-Diffusion-8B-BaseNVIDIA8.5B6.2 GBRuns well~91 tok/s413,039Fast
Ministral-3-8B-Instruct-2512Mistral AI8.9B6.3 GBRuns well~90 tok/s119,379Fast
granite-4.1-8bIbm8.8B6.3 GBRuns well~91 tok/s4,213,980Fast
Ornith-1.0-9BDeepreinforce9B6.7 GBRuns well~83 tok/s2,250,422Fast
Qwen3.5-9BAlibaba / Qwen9.7B6.8 GBRuns well~82 tok/s11,767,971Fast
GLM-4.6V-FlashZhipu AI10B7.0 GBRuns well~76 tok/s282,562Fast
Fara1.5-9BMicrosoft9.4B7.0 GBRuns well~79 tok/s14,526Fast
Olmo-3-7B-InstructAi27.3B7.2 GBRuns well~104 tok/s406,776Fast
gemma-4-12B-itGoogle DeepMind12B9.4 GBRuns well~65 tok/s2,963,990Fast
Ministral-3-14B-Instruct-2512Mistral AI14B9.7 GBRuns well~57 tok/s198,169Fast
Nemotron-Labs-Diffusion-14BNVIDIA14B9.7 GBRuns well~57 tok/s6,554Fast
Cosmos3-NanoNVIDIA16B11 GBRuns well~49 tok/s313,164Fast
ERNIE-4.5-21B-A3B-PTBaidu AI · MoE22B15 GBTight~253 tok/s3B active44,971Fast

Sorted by memory, smallest first — click any column to change it.

Upgrade

Including gemma-4-26B-A4B-it (27B), Qwen3.6-27B (28B) and Ornith-1.0-35B (35B) — around ~257 tok/s on a typical machine that size.

See everything 24GB VRAM runs →

All figures at Q4_K_M quantization, 4K context. “Memory” includes the model, its context cache and runtime overhead. Speed is calculated, not benchmarked: 717 GB/s × 0.65 ÷ bytes read per token, assuming a RTX 4080. Scale it by your own machine's bandwidth from the list above. Mixture-of-experts models read only their active parameters per token, which is why some large ones outrun smaller dense models.

Image generators that also fit

These make pictures rather than text. Their memory use scales with the resolution you generate at, not conversation length, so treat these figures as a floor rather than a ceiling.

What you'd need more memory for

The next models up, and what they ask for.

Other hardware

Verified against HuggingFace on 2026-08-08New to this? Start here →Runs on a laptop and up · last 12 months only