Skip to content
Local AI Wizard
Your machine
Theme

ThinkStation P3 Tiny Gen 2 · MTM 30K5S01D00

Lenovo · ThinkStation P3 Tiny Gen 2

Desktops & mini PCs8 GB VRAM + 32 GB system RAM

279 text models run well · 192 GB/s bandwidth

Models on this hardware

Start here

Text

Within 8 GB: 279 run well · 131 tight · 13 reviewed · 359 more run slowly (184 of them run well with 12 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 184 text models that would run well with 12 GB but not here already run here, slowly.
Text models that run well within 8 GB
Granite 4.2 3BIBM · 3B · LMArena 1,290Q4 · published params · source1,290 ± 12~5 GBQ4 · published paramssource
Granite 4.2 8BIBM · 8B · LMArena 1,290Q4 · published params · source1,290 ± 11~8 GBTight fit: Needs ~8 GB: runs well, nothing left beyond the system reserveQ4 · published paramssource
Llama 3.1 8B InstructMeta · 8B · LMArena 1,211Q4 · published params · source1,211 ± 4~8 GBTight fit: Needs ~8 GB: runs well, nothing left beyond the system reserveQ4 · 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
Gemma 4 E2BGoogle · 5.1BQ4 · published params · sourceNot on the LMArena text leaderboard of 2026-10-22~6 GBQ4 · published paramssource
Qwen3 4B Instruct 2507Alibaba · 4BQ4 · published params · sourceNot on the LMArena text leaderboard of 2026-10-22~5 GBQ4 · published paramssource
Qwen3 1.7BAlibaba · 1.7BQ4 · published params · sourceNot on the LMArena text leaderboard of 2026-10-22~4 GBQ4 · published paramssource
8 of 410

No model within 8 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 37

Every one that runs here, well or slowly, scored or not (13 run well, 24 slowly); 20 of 37 have a LiveBench or LMArena result. Run well first, then tight fits, then slow; within each, by Hub downloads over 30 days, read 2026-11-19, not by quality.

Reviewed text models that run on ThinkStation P3 Tiny Gen 2 · MTM 30K5S01D00
Gemma 4 E2BGoogle · 5.1BNot in LiveBench 2026-06-25Not on the LMArena text leaderboard of 2026-10-22~6 GB3.1Apache 2.02.8M
Qwen3 4B Instruct 2507Alibaba · 4BNot in LiveBench 2026-06-25Not on the LMArena text leaderboard of 2026-10-22~5 GB2.5Apache 2.02.8M
Qwen3 1.7BAlibaba · 1.7BNot in LiveBench 2026-06-25Not on the LMArena text leaderboard of 2026-10-22~4 GB1.1Apache 2.02.7M
Llama 3.2 1B InstructMeta · 1.2BNot in LiveBench 2026-06-251,111~3 GB0.8Llama 3.21.3M
Llama 3.2 3B InstructMeta · 3.2BNot in LiveBench 2026-06-251,167~5 GB2.0Llama 3.2823.8K
Phi-4 miniMicrosoft · 3.8BNot in LiveBench 2026-06-25Not on the LMArena text leaderboard of 2026-10-22~5 GB2.5MIT301.1K
SmolLM3 3BHugging Face · 3BNot in LiveBench 2026-06-25Not on the LMArena text leaderboard of 2026-10-22~5 GB1.9Apache 2.0185.3K
Ministral 3 3BMistral · 3.8BNot in LiveBench 2026-06-25Not on the LMArena text leaderboard of 2026-10-22~5 GB2.1Apache 2.072.5K
8 of 37
Also runs slowly335 models

Above the 8 GB of VRAM, the layers that do not fit run on the CPU from PC RAM, up to 32 GB in all (listed RAM: 32 GB). Planning estimate, not a speed figure. The 24 reviewed models that run slowly are in the table above.

6 of 335

Browse Text models

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

Vision

Within 8 GB: 666 run well · 44 tight · 2 reviewed · 338 more run slowly (207 of them run well with 12 GB) · 4 more with 12 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 8 GB
Phi-3.5-vision-instructmicrosoft · 4.15B · Image to text · LMArena 919Q4 · Hugging Face count · source919 ± 16~6 GBQ4 · Hugging Face countsource
Gemma 4 E2BGoogle · 5.1B · Image to textQ4 · published params · sourceNot on the LMArena vision leaderboard of 2026-10-19~6 GBQ4 · published paramssource
gemma-3n-E2Bgoogle · 5.44B · Image to textQ4 · Hugging Face count · sourceNot reviewed for LMArena~7 GBQ4 · Hugging Face countsource
gemma-3n-E2B-itgoogle · 5.44B · Image to textQ4 · Hugging Face count · sourceNot reviewed for LMArena~7 GBQ4 · Hugging Face countsource
metaclip-2-worldwide-huge-378facebook · 1.86B · Classic vision16-bit est. · Hugging Face count · sourceNot reviewed for LMArena~7 GB16-bit est. · Hugging Face countsource
metaclip-2-worldwide-huge-quickgelufacebook · 1.86B · Classic vision16-bit est. · Hugging Face count · sourceNot reviewed for LMArena~7 GB16-bit est. · Hugging Face countsource
Phi-4-multimodal-instructmicrosoft · 5.57B · Image to textQ4 · Hugging Face count · sourceNot on the LMArena vision leaderboard of 2026-10-19~7 GBQ4 · Hugging Face countsource
siglip2-giant-opt-patch16-256google · 1.87B · Classic vision16-bit est. · Hugging Face count · sourceNot reviewed for LMArena~7 GB16-bit est. · Hugging Face countsource
8 of 710

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 slowly338 models

Above the 8 GB of VRAM, the layers that do not fit run on the CPU from PC RAM, up to 32 GB in all (listed RAM: 32 GB). Planning estimate, not a speed figure.

6 of 338

With 12 GB you also run… 4

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

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

Image

74 within 8 GB · 15 more with 12 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 8 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
stable-diffusion-3-medium-diffusersstabilityai · 2.1B · Text to image16-bit est. · params · sourceNo comparable published resultNot reviewed for LMArena~8 GB16-bit est. · paramssource
stable-diffusion-xl-refiner-0.9stabilityai · 2.3B · Image to image16-bit est. · params · sourceNo comparable published resultNot reviewed for LMArena~8 GB16-bit est. · paramssource
stable-diffusion-xl-refiner-1.0stabilityai · 2.3B · Image to image16-bit est. · params · sourceNo comparable published resultNot reviewed for LMArena~8 GB16-bit est. · paramssource
URSA-1.7B-IBQ1024BAAI · 2.3B · Text to image16-bit est. · params · sourceNo comparable published resultNot reviewed for LMArena~8 GB16-bit est. · paramssource
URSA-1.7B-IBQ512BAAI · 2.3B · Text to image16-bit est. · params · sourceNo comparable published resultNot reviewed for LMArena~8 GB16-bit est. · paramssource
Lumina Image v2OpenGVLab · 2B · Text to image16-bit est. · params · sourceNo comparable published resultNot on the LMArena text-to-image leaderboard of 2026-10-07~7 GB16-bit est. · paramssource
8 of 74

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

With 12 GB you also run… 15

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

6 of 15

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

Video

40 within 8 GB · 9 more with 12 GB
Video models within 8 GB
Kandinsky-S2V-1.0-Lite-Diffuserskandinskylab · 2.3B · Image to video16-bit est. · params · source~8 GB16-bit est. · paramssource
Kandinsky-S2V-1.0-Lite-distill-Diffuserskandinskylab · 2.3B · Image to video16-bit est. · params · source~8 GB16-bit est. · paramssource
VACE-Wan2.1-1.3B-Previewali-vilab · 2.2B · Image to video16-bit est. · params · source~8 GB16-bit est. · paramssource
Wan2.1-VACE-1.3BWan-AI · 2.2B · Image to video16-bit est. · params · source~8 GB16-bit est. · paramssource
Wan2.1-VACE-1.3B-diffusersWan-AI · 2.2B · Image to video16-bit est. · params · source~8 GB16-bit est. · paramssource
CogVideoX-2bzai-org · 1.7B · Text to video16-bit est. · params · source~7 GB16-bit est. · paramssource
Kandinsky-WM-1.0-I2V-5s-AVkandinskylab · 2B · Image to video16-bit est. · params · source~7 GB16-bit est. · paramssource
Kandinsky-WM-1.0-I2V-5s-PHkandinskylab · 2B · Image to video16-bit est. · params · source~7 GB16-bit est. · paramssource
8 of 40

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

With 12 GB you also run… 9

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

6 of 9

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

Audio

241 within 8 GB · 55 more with 12 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 8 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
acestep-v15-baseACE-Step · 2.4B · Sound effects and audio16-bit est. · params · sourceNo comparable published result~8 GB16-bit est. · paramssource
acestep-v15-sftACE-Step · 2.4B · Sound effects and audio16-bit est. · params · sourceNo comparable published result~8 GB16-bit est. · paramssource
acestep-v15-turbo-continuousACE-Step · 2.4B · Sound effects and audio16-bit est. · params · sourceNo comparable published result~8 GB16-bit est. · paramssource
acestep-v15-turbo-shift1ACE-Step · 2.4B · Sound effects and audio16-bit est. · params · sourceNo comparable published result~8 GB16-bit est. · paramssource
acestep-v15-turbo-shift3ACE-Step · 2.4B · Sound effects and audio16-bit est. · params · sourceNo comparable published result~8 GB16-bit est. · paramssource
cohere-transcribe-03-2026CohereLabs · 2.1B · Speech to text16-bit est. · params · sourceNo comparable published result~8 GB16-bit est. · paramssource
8 of 241

With 12 GB you also run… 55

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

6 of 55

1,984 more audio models in the catalog have no usable memory figure or parameter count for this profile. Browse Audio models

Embeddings

585 run well within 8 GB · 49 more with 12 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 8 GB
Qwen3-Embedding 4BAlibaba (Qwen)8-bit · published params · source69.5~7 GB8-bit · published paramssource
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
8 of 585

With 12 GB you also run… 49

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

6 of 49

Browse Embeddings models

Runs well within 8 GB of VRAM · planning estimates, not measured speed live catalog · 19,444 repositories · checked

Measurements

Speed on this hardware

no published measurement yetWe have not yet included a verifiable measurement for this exact configuration. The memory estimates above do not indicate tokens per second or time to first token.

Specifications

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

GPU VRAM8 GBDedicated · what models load into · source
System RAM32 GBInstalled · not added to VRAM · source
Memory bandwidth192 GB/sManufacturer spec · not model speed · source
Where to buyLenovo.comManufacturer link · choose the configuration there · US · recorded

Marker: the lightest text model that does not run well here but runs well with 12 GB. No text model that runs well here has a LiveBench score. Q4 planning estimate, not a measured fit.

Reviewed · Manufacturer source

CPU cores
20
Processor
Intel Core Ultra 7 265T
Graphics processor
integrated Intel Graphics plus one NVIDIA RTX A1000
GPU runtime status
Not stated by the source
Memory capacity
32 GB
Memory kind
System RAM
Memory bandwidth
192 GB/s
CPU threads
20
CUDA cores
2,304
Dimensions (cm)
17.9 × 18.29 × 3.7 (W × D × H)
GPU VRAM
8 GB
GPU memory type
GDDR6
GPU outputs
4 × mini DisplayPort 1.4a
GPU power
50 W
Graphics type
Integrated Intel Graphics plus one discrete NVIDIA RTX A1000
Manufacturer part number
30K5S01D00
Memory interface
128 bit
Memory data rate
5,600 MT/s
Installed memory modules
1
Memory slots
2
System memory type
DDR5 SODIMM · non-ECC
Operating system
Windows 11 Pro · Traditional Chinese, Simplified Chinese, or English
Ports
USB-C 20 Gbps · USB-A 10 Gbps · USB-A 5 Gbps · DisplayPort 1.4 · HDMI 2.1 · RJ-45 Ethernet · 3.5 mm audio
Region
Not specified by the selected MTM page
Installed SSD capacity
1,024 GB
Storage options
Up to 3x M.2 SSD: PCIe 5.0 Performance up to 2 TB each; PCIe 4.0 Performance up to 1 TB each; PCIe 4.0 up to 1 TB each
Storage type
M.2 2280 PCIe 4.0 x4 Performance NVMe Opal
Introduced
2025

Family specification

Checked · Manufacturer source

NPU performance
13 TOPS · INT8

Lenovo PSREF model 30K5S01D00 lists a Core Ultra 7 265T, integrated Intel graphics, a separate RTX A1000 with 8 GB GDDR6, 32 GB installed DDR5-5600 system RAM, and a 1 TB SSD. System RAM and VRAM remain separate; runtime support and speed are unknown.

Photo: Lenovo family photo · memory configuration is not pictured. · Manufacturer source

Memory bandwidth is not measured model speed; verify runtime and checkpoint compatibility. Hardware specification reviewed: 2026-10-18.

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.