Skip to content
Local AI Wizard
Your machine
Theme

RTX 4050 Laptop GPU · 6 GB VRAM

NVIDIA · RTX 4050 Laptop GPU

41 text models run well · 192 GB/s bandwidth

Models on this hardware

Start here

Text

Within 6 GB: 41 run well · 7 tight · 9 reviewed · 111 more run slowly (29 of them run well with 8 GB)

Text · LMArena

rating · higher is better

  • Color = publisher

Source: LMArena 2026-10-22 · collected 2026-10-25 · dataset · CC BY 4.0 · Models without a rating are left out, not scored as zero.

Rows run well first, then tight fits; within each, reviewed models, then Hub downloads over 30 days, not quality. Sort a score column to rank by that source alone. All 29 text models that would run well with 8 GB but not here already run here, slowly.
Text models that run well within 6 GB
Granite 4.2 3BIBM · 3B · LMArena 1,290Q4 · published params · source1,290 ± 12~5 GBQ4 · published paramssource
Llama 3.2 3B InstructMeta · 3.21B · LMArena 1,167Q4 · published params · source1,167 ± 8~5 GBQ4 · published paramssource
Llama 3.2 1B InstructMeta · 1.23B · LMArena 1,111Q4 · published params · source1,111 ± 8~3 GBQ4 · published paramssource
Qwen3 1.7BAlibaba · 1.7BQ4 · published params · sourceNot on the LMArena text leaderboard of 2026-10-22~4 GBQ4 · published paramssource
Qwen3 4B Instruct 2507Alibaba · 4BQ4 · published params · sourceNot on the LMArena text leaderboard of 2026-10-22~5 GBQ4 · published paramssource
SmolLM3 3BHugging Face · 3BQ4 · published params · sourceNot on the LMArena text leaderboard of 2026-10-22~5 GBQ4 · published paramssource
Phi-4 miniMicrosoft · 3.8BQ4 · published params · sourceNot on the LMArena text leaderboard of 2026-10-22~5 GBQ4 · published paramssource
Ministral 3 3BMistral · 3.85BQ4 · tensor count · sourceNot on the LMArena text leaderboard of 2026-10-22~5 GBQ4 · tensor countsource
8 of 48

No model within 6 GB has a LiveBench score, so the table has no LiveBench column. LMArena ratings in this table: LMArena 2026-10-22 · collected 2026-10-25 · dataset · CC BY 4.0

Tight fit: it runs, but leaves less than 1 GB free beyond the 2 GB the system keeps (the memory figure already includes that reserve).

Every reviewed text model here 36

Every one that runs here, well or slowly, scored or not (9 run well, 27 slowly); 18 of 36 have a LiveBench or LMArena result. Run well first, then tight fits, then slow; within each, by Hub downloads over 30 days, read 2026-10-25, not by quality.

Reviewed text models that run on RTX 4050 Laptop GPU · 6 GB VRAM
Qwen3 1.7BAlibaba · 1.7BNot in LiveBench 2026-06-25Not on the LMArena text leaderboard of 2026-10-22~4 GB1.1Apache 2.04.1M
Llama 3.2 1B InstructMeta · 1.2BNot in LiveBench 2026-06-251,111~3 GB0.8Llama 3.23.5M
Qwen3 4B Instruct 2507Alibaba · 4BNot in LiveBench 2026-06-25Not on the LMArena text leaderboard of 2026-10-22~5 GB2.5Apache 2.02.6M
Llama 3.2 3B InstructMeta · 3.2BNot in LiveBench 2026-06-251,167~5 GB2.0Llama 3.21.1M
SmolLM3 3BHugging Face · 3BNot in LiveBench 2026-06-25Not on the LMArena text leaderboard of 2026-10-22~5 GB1.9Apache 2.0718.7K
Phi-4 miniMicrosoft · 3.8BNot in LiveBench 2026-06-25Not on the LMArena text leaderboard of 2026-10-22~5 GB2.5MIT342.8K
Ministral 3 3BMistral · 3.8BNot in LiveBench 2026-06-25Not on the LMArena text leaderboard of 2026-10-22~5 GB2.1Apache 2.087.4K
Granite 4.2 3BIBM · 3BNot in LiveBench 2026-06-251,290~5 GB2.2Apache 2.039.8K
8 of 36
Also runs slowly84 models

Above the 6 GB of VRAM, the layers that do not fit run on the CPU from PC RAM, up to 30 GB in all. 32 GB of PC RAM assumed: this configuration lists none. Planning estimate, not a speed figure. The 27 reviewed models that run slowly are in the table above.

6 of 84

Browse Text models

Adjust the estimate: quantization, context, reserveloads the live catalog

Vision

Within 6 GB: 56 run well · 21 tight · 1 reviewed · 64 more run slowly (6 of them run well with 8 GB) · 1 more with 8 GB

Vision · LMArena

rating · higher is better

Source: LMArena 2026-10-19 · collected 2026-10-25 · dataset · CC BY 4.0 · Models without a rating are left out, not scored as zero.

Rows run well first, then tight fits; within each, reviewed models, then Hub downloads over 30 days, not quality. Sort a score column to rank by that source alone.
Vision models that run well within 6 GB
Phi-3.5-vision-instructmicrosoft · 4.1B · Image to text · LMArena 919Q4 · Hugging Face count · source919 ± 16~6 GBTight fit: Needs ~6 GB: runs well, nothing left beyond the system reserveQ4 · Hugging Face countsource
CLIP-ViT-H-14-laion2B-s32B-b79Klaion · 0.99B · Classic vision16-bit est. · Hugging Face count · sourceNot on the LMArena vision leaderboard of 2026-10-19~5 GB16-bit est. · Hugging Face countsource
DeepSeek-OCRdeepseek-ai · 3.34B · Image to textQ4 · Hugging Face count · sourceNot on the LMArena vision leaderboard of 2026-10-19~5 GBQ4 · Hugging Face countsource
dots.mocrdots-studio · 3.04B · Image to textQ4 · Hugging Face count · sourceNot on the LMArena vision leaderboard of 2026-10-19~5 GBQ4 · Hugging Face countsource
dots.ocrdots-studio · 3.04B · Image to textQ4 · Hugging Face count · sourceNot on the LMArena vision leaderboard of 2026-10-19~5 GBQ4 · Hugging Face countsource
granite-4.0-3b-visionibm-granite · 4B · Image to textQ4 · Hugging Face count · sourceNot on the LMArena vision leaderboard of 2026-10-19~5 GBQ4 · Hugging Face countsource
granite-vision-3.3-2bibm-granite · 3B · Image to textQ4 · Hugging Face count · sourceNot on the LMArena vision leaderboard of 2026-10-19~5 GBQ4 · Hugging Face countsource
granite-vision-4.1-4bibm-granite · 4B · Image to textQ4 · Hugging Face count · sourceNot on the LMArena vision leaderboard of 2026-10-19~5 GBQ4 · Hugging Face countsource
8 of 77

LMArena ratings in this table: LMArena 2026-10-19 · collected 2026-10-25 · dataset · CC BY 4.0

Tight fit: it runs, but leaves less than 1 GB free beyond the 2 GB the system keeps (the memory figure already includes that reserve).

Also runs slowly64 models

Above the 6 GB of VRAM, the layers that do not fit run on the CPU from PC RAM, up to 30 GB in all. 32 GB of PC RAM assumed: this configuration lists none. Planning estimate, not a speed figure.

6 of 64

With 8 GB you also run… 1

The next memory size on this ladder runs these well. Figures are the requirement on this hardware type.

12 more vision models in the catalog have no usable memory figure or parameter count for this profile. Browse Vision models

Image

16 within 6 GB · 5 more with 8 GB

Image · Prompt alignment · GenEval

0–1 · higher is better

Source: GenEval · 2023 · reviewed 2026-10-05 · Published quality in the original benchmark preset; not measured on this hardware.

Image models within 6 GB
Stable Diffusion XL 1.0Stability AI · 2.6B · Text to imagePlatform path documented · stable-diffusion.cpp · source0.55Not on the LMArena text-to-image leaderboard of 2026-10-07Documented path without a memory figurePlatform path documentedstable-diffusion.cpp · source
Stable Diffusion 1.5Stability AI · 0.98B · Text to imagePlatform path documented · stable-diffusion.cpp · source0.43Not on the LMArena text-to-image leaderboard of 2026-10-07Documented path without a memory figurePlatform path documentedstable-diffusion.cpp · source
Sana Sprint 1.6BNVIDIA · 1.6B · Text to image16-bit est. · params · sourceNo comparable published resultNot on the LMArena text-to-image leaderboard of 2026-10-07~6 GB16-bit est. · paramssource
Sana_1600M_1024pxEfficient-Large-Model · 1.6B · Text to image16-bit est. · params · sourceNo comparable published resultNot reviewed for LMArena~6 GB16-bit est. · paramssource
TripoSGVAST-AI · 1.4B · 3D16-bit est. · params · sourceNo comparable published resultNot reviewed for LMArena~6 GB16-bit est. · paramssource
VGGT-1Bfacebook · 1.3B · 3D16-bit est. · params · sourceNo comparable published resultNot reviewed for LMArena~6 GB16-bit est. · paramssource
stable-diffusion-v1-4CompVis · 0.9B · Text to image16-bit est. · params · sourceNo comparable published resultNot reviewed for LMArena~5 GB16-bit est. · paramssource
stable-fast-3dstabilityai · 1B · 3D16-bit est. · params · sourceNo comparable published resultNot reviewed for LMArena~5 GB16-bit est. · paramssource
8 of 16

LMArena ratings in this table: LMArena 2026-10-07 · collected 2026-10-25 · dataset · CC BY 4.0

With 8 GB you also run… 5

The next memory size on this ladder. Figures are the requirement on this hardware type.

17 more image models in the catalog have no usable memory figure or parameter count for this profile. Browse Image models

Video

9 within 6 GB · 10 more with 8 GB
Video models within 6 GB
i2vgen-xlali-vilab · 1.4B · Text to video16-bit est. · params · source~6 GB16-bit est. · paramssource
stable-video-diffusion-img2vidstabilityai · 1.5B · Image to video16-bit est. · params · source~6 GB16-bit est. · paramssource
stable-video-diffusion-img2vid-xtstabilityai · 1.5B · Image to video16-bit est. · params · source~6 GB16-bit est. · paramssource
stable-video-diffusion-img2vid-xt-1-1stabilityai · 1.5B · Image to video16-bit est. · params · source~6 GB16-bit est. · paramssource
text-to-video-ms-1.7bali-vilab · 1.4B · Text to video16-bit est. · params · source~6 GB16-bit est. · paramssource
CogVideoX-5BZ.ai / Tsinghua KEG · 5B · Text to videoPlatform path documented · Diffusers / PyTorch · source~5 GBPlatform path documentedDiffusers / PyTorch · source
animatediff-motion-adapter-v1-5-2guoyww · 0.5B · Text to video16-bit est. · params · source~4 GB16-bit est. · paramssource
URSA-0.6B-FSQ320BAAI · 0.7B · Text to video16-bit est. · params · source~4 GB16-bit est. · paramssource
8 of 9

How this list works Documented execution paths on this hardware first, then publisher minimums (CUDA or Apple MPS).

With 8 GB you also run… 10

The next memory size on this ladder. Figures are the requirement on this hardware type.

6 of 10

20 more video models in the catalog have no usable memory figure or parameter count for this profile. Browse Video models

Audio

70 within 6 GB · 21 more with 8 GB

Audio · Word error rate · Open ASR

% · lower is better

Source: Word error rate · Open ASR · reviewed 2026-10-20 · Published quality in the original benchmark preset; not measured on this hardware.

Audio models within 6 GB
Whisper large-v3-turboOpenAI · 0.81B · Speech to textPlatform path documented · whisper.cpp / CUDA · source7.83Documented path without a memory figurePlatform path documentedwhisper.cpp / CUDA · source
Whisper large-v3OpenAI · 1.55B · Speech to textPlatform path documented · whisper.cpp / CUDA · source7.44Documented path without a memory figurePlatform path documentedwhisper.cpp / CUDA · source
Breeze-ASR-26MediaTek-Research · 1.5B · Speech to text16-bit est. · params · sourceNo comparable published result~6 GB16-bit est. · paramssource
csm-1bsesame · 1.6B · Text to speech16-bit est. · params · sourceNo comparable published result~6 GB16-bit est. · paramssource
Dia-1.6Bnari-labs · 1.6B · Text to speech16-bit est. · params · sourceNo comparable published result~6 GB16-bit est. · paramssource
LFM2.5-Audio-1.5BLiquidAI · 1.5B · Speech to speech / audio to audio16-bit est. · params · sourceNo comparable published result~6 GB16-bit est. · paramssource
Ming-omni-tts-0.5BinclusionAI · 1.6B · Text to speech16-bit est. · params · sourceNo comparable published result~6 GB16-bit est. · paramssource
musicgen-mediumfacebook · 1.5B · Sound effects and audio16-bit est. · params · sourceNo comparable published result~6 GB16-bit est. · paramssource
8 of 70

With 8 GB you also run… 21

The next memory size on this ladder. Figures are the requirement on this hardware type.

6 of 21

33 more audio models in the catalog have no usable memory figure or parameter count for this profile. Browse Audio models

Embeddings

69 run well within 6 GB · 6 more with 8 GB

Embeddings · MTEB (Multilingual, v2) · mean(task)

0–100 · higher is better

  • Color = publisher

Source: MTEB results (curated rows, sourced per model)

Embeddings models that run well within 6 GB
Harrier OSS v1 (0.6B)Microsoft8-bit · published params · source69.0~3 GB8-bit · published paramssource
Qwen3-Embedding 0.6BAlibaba (Qwen)8-bit · published params · source64.3~3 GB8-bit · published paramssource
Multilingual E5 Large Instructintfloat8-bit · published params · source63.2~3 GB8-bit · published paramssource
EmbeddingGemma 2Google8-bit · published params · source61.4~3 GB8-bit · published paramssource
EmbeddingGemma (300M)Google8-bit · published params · source61.1~3 GB8-bit · published paramssource
BGE-M3BAAI8-bit · published params · source59.6~3 GB8-bit · published paramssource
Nomic Embed Text v2 (MoE)Nomic AI8-bit · published params · source57.6~3 GB8-bit · published paramssource
Multilingual E5 Baseintfloat8-bit · published params · source57.0~3 GB8-bit · published paramssource
8 of 69

With 8 GB you also run… 6

The next memory size on this ladder runs these well. Figures are the requirement on this hardware type.

Browse Embeddings models

Runs well within 6 GB of VRAM · planning estimates, not measured speed bundled catalog copy · 1,777 repositories · checked

Measurements

Published tokens per second, exactly as measured; never estimated.

Published measurements on this hardware
Llama 2 7B Q4_0Llama 2 7B Q4_0 (scoreboard model; the scor…NVIDIA GeForce RTX 4050 Laptop · 6 GB GDDR643.72 ± 0.41Q4_0
0 / 128 · Single llama-bench process;…
CUDA
llama.cpp commit d79d8f3
Flash Attention disabled
llama.cpp CUDA scoreboard · community result
2026-10-25

Specifications

11 sourced fields; the source is named once per group, and a field from another source links to it. Official specifications

GPU VRAM6 GBDedicated · system RAM does not add · source
Memory bandwidth192 GB/sManufacturer spec · not model speed · source

Marker: the lightest text model that does not run well here but runs well with 8 GB. No text model that runs well here has a LiveBench score. Q4 planning estimate, not a measured fit.

Selected configuration

Latest check · Manufacturer source

Memory capacity
6 GB
Memory type
6 GB GDDR6 · 96-bit
Chip / processor
Ada Lovelace · RTX 4050 Laptop GPU
Memory bandwidth
192 GB/s

Family specification

Latest check · Manufacturer source

CUDA cores
2,560
GPU power (manufacturer specification; OEM may vary)
35–115 (OEM-dependent)
Memory interface
96 bit
Introduced
2023
Release date
2023-02-22
CUDA compute capability
8.9
Display connectors (reference design; OEM may vary)
HDMI 2.1a; DisplayPort 1.4a (system-dependent)

Ada laptop GPU with 6 GB GDDR6. NVIDIA lists a 35–115 W GPU subsystem range; each laptop chooses its power and cooling design. Treat the candidate list as a capacity estimate, not a compatibility or speed result.

Memory bandwidth is not measured model speed; verify runtime and checkpoint compatibility. Hardware specification reviewed: 2026-10-13.

How estimates work
Candidates
Reviewed pages use published parameter counts; automatic text candidates use official repository tensor counts (packed quantized weights and unsupported architectures excluded).
Planning
Q4 weights + 20% allowance + reserve: 25% of unified or system memory (at least 2 GB, at most 8 GB; 2 GB here) or 2 GB per dedicated GPU. Adjustable in “Adjust the estimate”. No KV cache is calculated from context.
Fit tiers
Runs well: the estimate fits the fast memory (VRAM, or unified memory within the macOS GPU share). Runs slowly: within the machine's memory but outside the fast part; a graphics card offloads layers to PC RAM (the listed RAM, or 32 GB assumed). Doesn't run: above both. Tiers are planning estimates, never a speed.
Vision
Vision-language models use the Q4 planning estimate from the parameter count (published by the model page, or read from Hugging Face). Classic vision models (classify, detect, segment) use a 16-bit estimate. No vision benchmark is tracked, so rows carry no score.
Media and embeddings
Media rows use documented execution paths (exact hardware or platform) and publisher minimums; a published requirement above this memory moves a model to the next tier. Without either, a 16-bit estimate from the published parameter count (× 1.2 + 2 GB), labelled as such. Embeddings use 8-bit weights + 20% + 2 GB.

Local AI basics: parameters, Q4 and memory · Full model catalog · Data and methodology

Alternatives

Other machines on the same memory ladder (VRAM); each page lists the models that fit it.