GPU Vault model library

Launch text-generation models on dedicated GPUs.

Find validated models for chat, coding, summarization, extraction, and other private text workloads.

Sort

Text-generation models

14 launchable models on this page. Exact, runtime-qualified revisions only.

text generationmeta-llama

Llama-3.2-3B-Instruct

Llama-3.2-3B-Instruct is a validated text generation model.

1× GPU · 24 GB VRAM3.2B params 1.9M

text generationQwen

Qwen2.5-Coder-14B-Instruct

Qwen2.5-Coder-14B-Instruct is a validated text generation model.

1× GPU · 48 GB VRAM14.8B params 1.8M

text generationHuggingFaceTB

SmolLM2-135M

SmolLM2-135M is a validated text generation model.

1× GPU · 16 GB VRAM134.5M params 1.8M

text generationmistralai

Mistral-7B-Instruct-v0.2

Mistral-7B-Instruct-v0.2 is a validated text generation model.

1× GPU · 40 GB VRAM7.2B params 1.8M

text generationHuggingFaceTB

SmolLM2-135M-Instruct

SmolLM2-135M-Instruct is a validated text generation model.

1× GPU · 16 GB VRAM134.5M params 1.7M

text generationQwen

Qwen3-30B-A3B

Qwen3-30B-A3B is a validated text generation model.

1× GPU · 80 GB VRAM30.5B params 1.6M

text generationQwen

Qwen2.5-0.5B

Qwen2.5-0.5B is a validated text generation model.

1× GPU · 16 GB VRAM494M params 1.6M

text generationQwen

Qwen3-1.7B-Base

Qwen3-1.7B-Base is a validated text generation model.

1× GPU · 24 GB VRAM1.7B params 1.5M

text generationQwen

Qwen3-4B-Base

Qwen3-4B-Base is a validated text generation model.

1× GPU · 24 GB VRAM4B params 1.5M

text generationQwen

Qwen2.5-Coder-32B-Instruct-AWQ

Qwen2.5-Coder-32B-Instruct-AWQ is a validated text generation model.

2× GPU · 80 GB VRAM32.8B params 1.4M

text generationTinyLlama

TinyLlama-1.1B-Chat-v1.0

TinyLlama-1.1B-Chat-v1.0 is a validated text generation model.

1× GPU · 16 GB VRAM1.1B params 1.4M

text generationEleutherAI

pythia-70m-deduped

pythia-70m-deduped is a validated text generation model.

1× GPU · 16 GB VRAM95.6M params 1.3M

text generationtrl-internal-testing

tiny-Qwen3ForCausalLM

tiny-Qwen3ForCausalLM is a validated text generation model.

1× GPU · 16 GB VRAM2.5M params 1.2M

text generationmeta-llama

Meta-Llama-3-8B-Instruct

Meta-Llama-3-8B-Instruct is a validated text generation model.

1× GPU · 40 GB VRAM8B params 1.2M