Qwen's 2400B-parameter mixture-of-experts (95B active) open language model with a 256K-token context, ready to serve across a multi-node Runpod Cluster.

Language

2.4T MoE · 95B active · 256K context

Text generation

Qwen/Qwen3.8-2.4T-A95B

2026-08-07 17:00

Clusters

Qwen

Qwen3.8 27B

Qwen's 27B-parameter open vision-language model with a 256K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Unlimited OCR

Baidu's 3.3B-parameter open vision-language model with a 32K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Qwen3.6 35B A3B

Qwen's 35B-parameter mixture-of-experts (3B active) open vision-language model with a 256K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Vision + language

35B MoE · 3B active · 256K context

Gemma 4 31B IT

Google's 31B-parameter open vision-language model with a 256K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Vision + language

31B · 256K context

Text generation

google/gemma-4-31B-it

LocateAnything 3B

NVIDIA's 3B-parameter open vision-language model with a 32K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Vision + language

3B · 32K context

Text generation

nvidia/LocateAnything-3B

Qwen3.5 4B

Qwen's 4B-parameter open vision-language model with a 256K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Qwen3.5 9B

Qwen's 9B-parameter open vision-language model with a 256K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

IBM Granite 4.0 H Small

IBM Granite-4.0-H-Small is a 32B parameter long-context instruct model.

Language

Text generation

ibm-granite/granite-4.0-h-small

2025-10-07 17:00

Public Endpoint

Qwen

Qwen3 32B AWQ

Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.

GPT-OSS 120B

OpenAI's 117B-parameter mixture-of-experts (5.1B active) open language model with a 128K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Language

117B MoE · 5.1B active · 128K context

GPT-OSS 20B

OpenAI's 21B-parameter mixture-of-experts (3.6B active) open language model with a 128K-token context, ready to serve on Runpod Serverless with vLLM or on a Pod.

Language

21B MoE · 3.6B active · 128K context

Llama 3.2 3B Instruct

Meta's 3B-parameter open language model, ready to serve on Runpod Serverless with vLLM or on a Pod.

Language

Text generation

meta-llama/Llama-3.2-3B-Instruct

Seedance 2.5 Image-to-Video

ByteDance Seedance 2.5 animates a start image (and optional last frame) into a 4-30 second video up to 4K, driven by a text prompt, with native audio.

Video

Image to video

bytedance/seedance-2.5

2026-10-12 17:00

Public Endpoint

ByteDance

Seedance 2.5 Text-to-Video

ByteDance Seedance 2.5 generates 4-30 second videos up to 4K from a text prompt, with native audio and optional reference images, videos and audio.

Video

Text to video

bytedance/seedance-2.5

2026-10-12 17:00

Public Endpoint

Pruna

P-Video-Edit

Edit an existing video from a text prompt, optionally guided by up to 4 reference images, by Pruna. Input videos up to 15 seconds.

Hailuo 2.3 Fast

MiniMax's Hailuo 2.3 Fast model for AI video generation. Animate a reference image with a text prompt to generate 6 or 10 second videos with automatic prompt enhancement.

Hailuo 02 Standard

MiniMax's Hailuo 02 Standard model for AI video generation. Generate 6 or 10 second videos from text prompts or reference images with automatic prompt enhancement.

Veo 3.1 Fast Image-to-Video

Google's Veo 3.1 Fast model for high-quality image-to-video generation. Animate a reference image with a text prompt at up to 4K resolution with optional audio.

Nano Banana 2

Google's Nano Banana 2 model for AI-powered image editing. Provide up to 14 reference images and a text prompt to produce edited outputs at 1K, 2K, or 4K resolution.

P-Video

Premium AI video generation from text, images, and audio by Pruna. Generation in less than 10 seconds.

Vidu Q3 Image-to-Video

Vidu Q3 animates a reference image into a video with motion driven by a text prompt. Supports multiple resolutions up to 1080p, adjustable duration, and optional audio generation.

Vidu Q3 Text-to-Video

Vidu Q3 generates high-quality videos from text descriptions with support for multiple resolutions, aspect ratios, and audio generation.

Z Image Turbo

Z-Image is a powerful and highly efficient image generation model family with 6B parameters

Kling v2.6 Standard Motion Control

Transfers motion from reference videos to animate still images. Upload a character image and a motion clip, the model extracts the movement to generate smooth video

Qwen Image Edit 2511 LoRA

Qwen series that achieves significant advances in complex text rendering and precise image editing and has LoRA support.

Qwen Image Edit 2511

It delivers stronger edit consistency, robust multi-person identity/pose consistency, built-in LoRA styles, enhanced industrial/product design.

WAN 2.6 T2I

Alibaba WAN 2.6 Text-to-Image generates high-quality images from natural-language prompts with strong prompt adherence and clean composition.

WAN 2.6 T2V

Alibaba WAN 2.6 Text-to-Video turns plain prompts into coherent, cinematic clips with crisp detail, stable motion, and strong instruction-following

Seedance V1.5 Pro I2V

Generates cinematic, live-action–leaning clips from a text prompt, preserving the image’s subject and composition while adding expressive motion and stable aesthetics.

Kling Video O1 R2V

Kling video generates creative videos using character, prop, or scene references from multiple viewpoints.

Wan 2.6 I2V

Alibaba WAN 2.6 is a state-of-the-art image-to-video generation model available.

Chatterbox Turbo

Resemble AI's Chatterbox Turbo is the fastest open source text to speech model with paralinguistic tags for non-speech sounds with expressive voice cloning capabilities.

Audio

Text to speech

resembleai/chatterbox-turbo

2025-12-14 16:00

Public Endpoint

Pruna

Pruna Image Edit

P-Image-Edit is Pruna's premium image editing model. It supports complex compositions, style transfers, and targeted edits with text instructions.

Pruna Image T2I

P-Image is Pruna's ultra-fast text-to-image model with automatic prompt enhancement and 2-stage refinement.

Nano Banana Pro Edit

Google Nano Banana Pro (Gemini 3.0 Pro Image) Edit enables image editing with 4K-capable output.

SORA 2 Pro I2V

OpenAI's Sora 2 Pro is new state of the art video and audio generation model.

InfiniteTalk

InfiniteTalk is an audio-driven conversational AI video generation model. Create talking or singing videos from a single image and audio input.

SORA 2 I2V

OpenAI's Sora 2 is new state of the art video and audio generation model.

Wan 2.5 I2V

Alibaba WAN 2.5 is a state-of-the-art image-to-video generation model available.

Seedream 4.0 T2I

As a new-generation image creation model, Seedream 4.0 integrates image generation and image editing capabilities into a single, unified architecture.

Seedream 4.0 Edit

As a new-generation image creation model, Seedream 4.0 integrates image generation and image editing capabilities into a single, unified architecture.

Wan 2.2 I2V 720p LoRA

Wan 2.2 is an open-source AI video generation model that utilizes a diffusion transformer architecture and a novel 3D spatio-temporal VAE (Wan-VAE) for image-to-video generation.

Qwen Image Edit

The image editing version of Qwen-Image. Qwen-Image-Edit successfully extends Qwen-Image’s unique text rendering capabilities to image editing tasks, enabling precise text editing.

Minimax Speech 02 HD

MiniMax is a high-definition text-to-speech model

Qwen Image LoRA

An image generation foundation model in the Qwen series that achieves significant advances in complex text rendering and precise image editing and has LoRA support.

Qwen Image

an image generation foundation model in the Qwen series that achieves significant advances in complex text rendering and precise image editing.

Wan 2.2 I2V 720p

Wan 2.2 is an open-source AI video generation model that utilizes a diffusion transformer architecture and a novel 3D spatio-temporal VAE (Wan-VAE) for image-to-video generation.

FLUX.1 Kontext [dev]

FLUX.1 Kontext [dev] is a 12 billion parameter rectified flow transformer capable of editing images based on text instructions.

Wan 2.1 I2V 720p

Wan 2.1 is an open-source AI video generation model that utilizes a diffusion transformer architecture and a novel 3D spatio-temporal VAE (Wan-VAE) for image-to-video generation.

FLUX.1 [dev]

Offers exceptional prompt adherence, high visual fidelity, and rich image detail.

FLUX.1 Schnell

Fastest and most lightweight FLUX model, ideal for local development, prototyping, and personal use.

How should I deploy an AI model?

Use Public Endpoints for API access to pre-deployed models. Choose Serverless for your own inference endpoint, Pods for control of a dedicated environment, or Clusters for workloads that span multiple nodes.

Compare Public Endpoints, Serverless, Pods, and Clusters
Deployment option	Choose it when	Setup and control	Scaling	How billing works
Public Endpoints	Call a supported, pre-deployed model through an API.	Use a Runpod API key. Runpod hosts the model.	Runpod manages the model infrastructure.	Usage-based rates vary by model and output type.
Serverless	Serve your own model behind an endpoint.	Deploy a worker image and configure your endpoint.	Flex workers can scale to zero; active workers stay running.	Compute is billed per second while workers run, including startup and idle timeout. Storage is separate.
Pods	Use a custom serving stack, fine-tune, or run persistent workloads.	Configure a dedicated environment and manage the software.	Manage your Pod resources directly.	Compute is billed per second while the Pod runs. Retained storage can still incur charges.
Clusters	Run distributed training or inference across multiple nodes.	Configure your workload on a managed multi-node cluster.	Choose the node configuration for your workload.	Rates depend on GPU type and node count. See current pricing in the linked documentation.

Deployment options vary by model. Check the model page for supported routes, then follow the linked documentation for setup and billing details.

Deploy any open model from Hugging Face

Runpod Serverless serves open models with vLLM. Pick one of these 314 for a page with one-click deploy, or paste any Hugging Face model ID when you create an endpoint.

Create a Serverless endpoint vLLM worker docs

agentica-org/deepcoder-14b-preview with a custom API endpoint

agentica-org/deepcoder-1.5b-preview with a custom API endpoint

agentica-org/deepscaler-1.5b-preview with a custom API endpoint

aidc-ai/marco-o1 with a custom API endpoint

ai-mo/kimina-autoformalizer-7b with a custom API endpoint

ai-mo/kimina-prover-preview-distill-1.5b with a custom API endpoint

ai-mo/kimina-prover-preview-distill-7b with a custom API endpoint

aiteamvn/grpo-vi-qwen2-7b-rag with a custom API endpoint

aixonlab/eurydice-24b-v2 with a custom API endpoint

alamios/mistral-small-3.1-draft-0.5b with a custom API endpoint

allam-ai/allam-7b-instruct-preview with a custom API endpoint

allenai/llama-3.1-tulu-3-8b with a custom API endpoint

allenai/olmo-2-0425-1b-instruct with a custom API endpoint

all-hands/openhands-lm-1.5b-v0.1 with a custom API endpoint

all-hands/openhands-lm-7b-v0.1 with a custom API endpoint

almawave/velvet-14b with a custom API endpoint

arcee-ai/arcee-blitz with a custom API endpoint

arcee-ai/arcee-maestro-7b-preview with a custom API endpoint

arcee-ai/virtuoso-lite with a custom API endpoint

arcee-ai/virtuoso-small-v2 with a custom API endpoint

arliai/qwq-32b-arliai-rpr-v4 with a custom API endpoint

arshiaafshani/arshgpt with a custom API endpoint

arshiaafshani/arsh-llm with a custom API endpoint

arshiaafshani/arshstory with a custom API endpoint

atlaai/selene-1-mini-llama-3.1-8b with a custom API endpoint

axcxept/phi-4-deepseek-r1k-rl-ezo with a custom API endpoint

axcxept/phi-4-open-r1-distill-ezov1 with a custom API endpoint

bespokelabs/bespoke-stratos-7b with a custom API endpoint

bigcode/starcoder with a custom API endpoint

bllossom/llama-3.2-korean-bllossom-3b with a custom API endpoint

bsc-lt/salamandra-7b with a custom API endpoint

bytedance-research/bfs-prover with a custom API endpoint

bytedance-seed/seed-coder-8b-base with a custom API endpoint

bytedance-seed/seed-coder-8b-instruct with a custom API endpoint

bytedance-seed/seed-coder-8b-reasoning-bf16 with a custom API endpoint

cognitivecomputations/dolphin3.0-mistral-24b with a custom API endpoint

cognitivecomputations/dolphin3.0-r1-mistral-24b with a custom API endpoint

contactdoctor/bio-medical-llama-3-8b with a custom API endpoint

cyberagent/deepseek-r1-distill-qwen-14b-japanese with a custom API endpoint

darkc0de/xortroncriminalcomputingconfig with a custom API endpoint

davanstrien/smol-hub-tldr with a custom API endpoint

deepcogito/cogito-v1-preview-llama-3b with a custom API endpoint

deepcogito/cogito-v1-preview-llama-8b with a custom API endpoint

deepcogito/cogito-v1-preview-qwen-14b with a custom API endpoint

deepmount00/llama-3.1-8b-ita with a custom API endpoint

deepseek-ai/deepseek-coder-6.7b-instruct with a custom API endpoint

deepseek-ai/deepseek-llm-7b-base with a custom API endpoint

deepseek-ai/deepseek-llm-7b-chat with a custom API endpoint

deepseek-ai/deepseek-r1-distill-llama-8b with a custom API endpoint

deepseek-ai/deepseek-r1-distill-qwen-14b with a custom API endpoint

deepseek-ai/deepseek-r1-distill-qwen-1.5b with a custom API endpoint

deepseek-ai/deepseek-r1-distill-qwen-7b with a custom API endpoint

defog/sqlcoder-7b-2 with a custom API endpoint

delta-vector/rei-12b with a custom API endpoint

delta-vector/rei-v2-12b with a custom API endpoint

distilbert/distilgpt2 with a custom API endpoint

dnotitia/dna-r1 with a custom API endpoint

dreamgen/lucid-v1-nemo with a custom API endpoint

driaforall/dria-agent-a-7b with a custom API endpoint

efficientscaling/z1-7b with a custom API endpoint

facebook/kernelllm with a custom API endpoint

fdtn-ai/foundation-sec-8b with a custom API endpoint

fluently-lm/fluentlylm-prinum with a custom API endpoint

fractalairesearch/fathom-r1-14b with a custom API endpoint

goppa-ai/goppa-logillama with a custom API endpoint

gryphe/mythomax-l2-13b with a custom API endpoint

hoangha/pensez-v0.1-e5 with a custom API endpoint

homebrewltd/alphamaze-v0.2-1.5b with a custom API endpoint

huggingfaceh4/zephyr-7b-beta with a custom API endpoint

huggingfacetb/smollm2-135m-instruct with a custom API endpoint

huggingfacetb/smollm2-135m with a custom API endpoint

huggingfacetb/smollm2-1.7b-instruct with a custom API endpoint

huggingfacetb/smollm2-360m-instruct with a custom API endpoint

ibm-granite/granite-3.1-8b-instruct with a custom API endpoint

ibm-granite/granite-3.2-2b-instruct with a custom API endpoint

ibm-granite/granite-3.2-8b-instruct-preview with a custom API endpoint

ibm-granite/granite-3.2-8b-instruct with a custom API endpoint

ibm-granite/granite-3.3-2b-instruct with a custom API endpoint

ibm-granite/granite-3.3-8b-base with a custom API endpoint

ibm-granite/granite-3.3-8b-instruct with a custom API endpoint

ihor/text2graph-r1-qwen2.5-0.5b with a custom API endpoint

iic/rigochat-7b-v2 with a custom API endpoint

ilsp/llama-krikri-8b-base with a custom API endpoint

ilsp/llama-krikri-8b-instruct with a custom API endpoint

inceptionai/llama-3.1-sherkala-8b-chat with a custom API endpoint

internlm/oreal-7b with a custom API endpoint

jetbrains/mellum-4b-base with a custom API endpoint

jetbrains/mellum-4b-sft-kotlin with a custom API endpoint

jinaai/reader-lm-1.5b with a custom API endpoint

jinaai/readerlm-v2 with a custom API endpoint

kakaocorp/kanana-1.5-2.1b-instruct-2505 with a custom API endpoint

kakaocorp/kanana-1.5-8b-base with a custom API endpoint

kakaocorp/kanana-1.5-8b-instruct-2505 with a custom API endpoint

kakaocorp/kanana-nano-2.1b-base with a custom API endpoint

kakaocorp/kanana-nano-2.1b-instruct with a custom API endpoint

kakaocorp/kanana-safeguard-8b with a custom API endpoint

kakaocorp/kanana-safeguard-prompt-2.1b with a custom API endpoint

katanemo/arch-function-3b with a custom API endpoint

kblueleaf/tipo-500m-ft with a custom API endpoint

knifeayumu/cydonia-v1.3-magnum-v4-22b with a custom API endpoint

knoveleng/open-rs3 with a custom API endpoint

kyutai/helium-1-2b with a custom API endpoint

kz919/qwq-0.5b-distilled-sft with a custom API endpoint

latitudegames/muse-12b with a custom API endpoint

latitudegames/wayfarer-12b with a custom API endpoint

lgai-exaone/exaone-3.5-2.4b-instruct with a custom API endpoint

lgai-exaone/exaone-deep-2.4b with a custom API endpoint

lgai-exaone/exaone-deep-32b with a custom API endpoint

lgai-exaone/exaone-deep-7.8b with a custom API endpoint

lightblue/deepseek-r1-distill-qwen-7b-japanese with a custom API endpoint

lightblue/lb-reranker-0.5b-v1.0 with a custom API endpoint

livekit/turn-detector with a custom API endpoint

llm-jp/llm-jp-3.1-13b-instruct4 with a custom API endpoint

locutusque/thespis-llama-3.1-8b with a custom API endpoint

malteos/german-r1 with a custom API endpoint

m-a-p/yue-s1-7b-anneal-en-cot with a custom API endpoint

m-a-p/yue-s1-7b-anneal-en-icl with a custom API endpoint

m-a-p/yue-s1-7b-anneal-jp-kr-cot with a custom API endpoint

m-a-p/yue-s1-7b-anneal-zh-cot with a custom API endpoint

m-a-p/yue-s2-1b-general with a custom API endpoint

marin-community/marin-8b-instruct with a custom API endpoint

menlo/rezero-v0.1-llama-3.2-3b-it-grpo-250404 with a custom API endpoint

meta-llama/codellama-7b-hf with a custom API endpoint

meta-llama/llama-2-7b-hf with a custom API endpoint

meta-llama/llama-3.1-8b-instruct with a custom API endpoint

meta-llama/llama-3.2-1b-instruct with a custom API endpoint

meta-llama/llama-3.2-3b with a custom API endpoint

meta-llama/llama-guard-3-8b with a custom API endpoint

meta-llama/meta-llama-3-8b-instruct with a custom API endpoint

meta-llama/meta-llama-3-8b with a custom API endpoint

microsoft/dialogpt-medium with a custom API endpoint

microsoft/phi-2 with a custom API endpoint

microsoft/phi-3.5-mini-instruct with a custom API endpoint

microsoft/phi-3-mini-4k-instruct with a custom API endpoint

microsoft/phi-4-reasoning-plus with a custom API endpoint

microsoft/phi-4 with a custom API endpoint

mistralai/codestral-22b-v0.1 with a custom API endpoint

mistralai/mistral-7b-instruct-v0.1 with a custom API endpoint

mistralai/mistral-7b-instruct-v0.2 with a custom API endpoint

mistralai/mistral-7b-v0.1 with a custom API endpoint

mistralai/mistral-7b-v0.3 with a custom API endpoint

mistralai/mistral-small-24b-base-2501 with a custom API endpoint

mistralai/mistral-small-24b-instruct-2501 with a custom API endpoint

mixedbread-ai/mxbai-rerank-base-v2 with a custom API endpoint

mixedbread-ai/mxbai-rerank-large-v2 with a custom API endpoint

mlp-ktlim/llama-3-korean-bllossom-8b with a custom API endpoint

mobiuslabsgmbh/deepseek-r1-redistill-qwen-1.5b-v1.0 with a custom API endpoint

mobiuslabsgmbh/deepseek-r1-redistill-qwen-7b-v1.1 with a custom API endpoint

modelcloud/qwq-32b-preview-gptqmodel-4bit-vortex-v3 with a custom API endpoint

mrfakename/mistral-small-3.1-24b-instruct-2503-hf with a custom API endpoint

naver-hyperclovax/hyperclovax-seed-text-instruct-0.5b with a custom API endpoint

navid-ai/yehia-7b-preview with a custom API endpoint

netease-youdao/confucius-o1-14b with a custom API endpoint

nexaaidev/octo-net with a custom API endpoint

nousresearch/deephermes-3-llama-3-3b-preview with a custom API endpoint

nousresearch/deephermes-3-llama-3-8b-preview with a custom API endpoint

nousresearch/deephermes-3-mistral-24b-preview with a custom API endpoint

nousresearch/hermes-3-llama-3.1-8b with a custom API endpoint

nousresearch/hermes-3-llama-3.2-3b with a custom API endpoint

nousresearch/nous-hermes-2-mistral-7b-dpo with a custom API endpoint

numind/nuextract-1.5 with a custom API endpoint

nvidia/aceinstruct-1.5b with a custom API endpoint

nvidia/acemath-7b-instruct with a custom API endpoint

nvidia/acemath-rl-nemotron-7b with a custom API endpoint

nvidia/acereason-nemotron-14b with a custom API endpoint

nvidia/acereason-nemotron-7b with a custom API endpoint

nvidia/llama-3.1-nemotron-nano-4b-v1.1 with a custom API endpoint

nvidia/llama-3.1-nemotron-nano-8b-v1 with a custom API endpoint

nvidia/opencodereasoning-nemotron-14b with a custom API endpoint

nvidia/opencodereasoning-nemotron-7b with a custom API endpoint

nvidia/openmath-nemotron-14b-kaggle with a custom API endpoint

nvidia/openmath-nemotron-14b with a custom API endpoint

nvidia/openmath-nemotron-1.5b with a custom API endpoint

nvidia/openmath-nemotron-7b with a custom API endpoint

openai-community/gpt2 with a custom API endpoint

open-neo/kyro-n1-3b with a custom API endpoint

open-r1/olympiccoder-7b with a custom API endpoint

open-r1/openr1-distill-7b with a custom API endpoint

open-thoughts/openthinker2-7b with a custom API endpoint

open-thoughts/openthinker-7b with a custom API endpoint

oumi-ai/halloumi-8b with a custom API endpoint

ozone-ai/0x-lite with a custom API endpoint

ozone-ai/reverb-7b with a custom API endpoint

ozone-research/reverb-7b with a custom API endpoint

pku-ds-lab/fairyr1-14b-preview with a custom API endpoint

pocketdoc/dans-personalityengine-v1.2.0-24b with a custom API endpoint

pocketdoc/dans-personalityengine-v1.3.0-12b with a custom API endpoint

pocketdoc/dans-personalityengine-v1.3.0-24b with a custom API endpoint

powerinfer/smallthinker-3b-preview with a custom API endpoint

prime-rl/eurus-2-7b-prime with a custom API endpoint

prithivmlmods/calcium-opus-14b-elite2-r1 with a custom API endpoint

prithivmlmods/calcium-opus-14b-elite with a custom API endpoint

prithivmlmods/calcium-opus-20b-v1 with a custom API endpoint

prithivmlmods/coma-ii-14b with a custom API endpoint

prithivmlmods/cygnus-ii-14b with a custom API endpoint

prithivmlmods/deepthink-llama-3-8b-preview with a custom API endpoint

prithivmlmods/deepthink-reasoning-14b with a custom API endpoint

prithivmlmods/dinobot-opus-14b-exp with a custom API endpoint

prithivmlmods/epimetheus-14b-axo with a custom API endpoint

prithivmlmods/equuleus-opus-14b-exp with a custom API endpoint

prithivmlmods/eridanus-opus-14b-r999 with a custom API endpoint

prithivmlmods/evac-opus-14b-exp with a custom API endpoint

prithivmlmods/gaea-opus-14b-exp with a custom API endpoint

prithivmlmods/galactic-qwen-14b-exp1 with a custom API endpoint

prithivmlmods/galactic-qwen-14b-exp2 with a custom API endpoint

prithivmlmods/gauss-opus-14b-r999 with a custom API endpoint

prithivmlmods/llama-3.2-6b-algocode with a custom API endpoint

prithivmlmods/llama-8b-distill-cot with a custom API endpoint

prithivmlmods/lwq-reasoner-10b with a custom API endpoint

prithivmlmods/magellanic-opus-14b-exp with a custom API endpoint

prithivmlmods/magellanic-qwen-25b-r999 with a custom API endpoint

prithivmlmods/megatron-opus-14b-2.1 with a custom API endpoint

prithivmlmods/megatron-opus-14b-exp with a custom API endpoint

prithivmlmods/messier-opus-14b-elite7 with a custom API endpoint

prithivmlmods/omni-reasoner2-merged with a custom API endpoint

prithivmlmods/omni-reasoner3-merged with a custom API endpoint

prithivmlmods/pegasus-opus-14b-exp with a custom API endpoint

prithivmlmods/phi-4-o1 with a custom API endpoint

prithivmlmods/phi-4-super-1 with a custom API endpoint

prithivmlmods/phi-4-super with a custom API endpoint

prithivmlmods/porpoise-opus-14b-exp with a custom API endpoint

prithivmlmods/primal-mini-3b-exp with a custom API endpoint

prithivmlmods/primal-opus-14b-optimus-v1 with a custom API endpoint

prithivmlmods/primal-opus-14b-optimus-v2 with a custom API endpoint

prithivmlmods/qwq-lcot-14b-conversational with a custom API endpoint

prithivmlmods/qwq-lcot1-merged with a custom API endpoint

prithivmlmods/qwq-lcot2-7b-instruct with a custom API endpoint

prithivmlmods/qwq-math-io-500m with a custom API endpoint

prithivmlmods/qwq-r1-distill-7b-cot with a custom API endpoint

prithivmlmods/qwq-supernatural-3b with a custom API endpoint

prithivmlmods/smollm2_135m_grpo_checkpoint with a custom API endpoint

prithivmlmods/smollm2_135m_grpo_gsm8k with a custom API endpoint

prithivmlmods/smollm2-360m-grpo-r999 with a custom API endpoint

prithivmlmods/sombrero-opus-14b-elite5 with a custom API endpoint

prithivmlmods/sombrero-opus-14b-elite6 with a custom API endpoint

prithivmlmods/sombrero-opus-14b-sm1 with a custom API endpoint

prithivmlmods/sombrero-opus-14b-sm2 with a custom API endpoint

prithivmlmods/sombrero-opus-14b-sm4 with a custom API endpoint

prithivmlmods/sombrero-opus-14b-sm5 with a custom API endpoint

prithivmlmods/sqweeks-7b-instruct with a custom API endpoint

prithivmlmods/taurus-opus-7b with a custom API endpoint

prithivmlmods/tucana-opus-14b-r999 with a custom API endpoint

prithivmlmods/viper-coder-hybridmini-v1.3 with a custom API endpoint

prithivmlmods/viper-coder-hybrid-v1.2 with a custom API endpoint

prithivmlmods/viper-coder-hybrid-v1.3 with a custom API endpoint

prithivmlmods/viper-coder-v1.1 with a custom API endpoint

prithivmlmods/viper-coder-v1.6-r999 with a custom API endpoint

prithivmlmods/viper-onecoder-uigen with a custom API endpoint

prithivmlmods/volans-opus-14b-exp with a custom API endpoint

probemedicalyonseimailab/medllama3-v20 with a custom API endpoint

qihoo360/light-r1-14b-ds with a custom API endpoint

qihoo360/light-r1-32b with a custom API endpoint

qihoo360/light-r1-7b-ds with a custom API endpoint

qwen/qwen2.5-0.5b-instruct with a custom API endpoint

qwen/qwen2.5-0.5b with a custom API endpoint

qwen/qwen2.5-14b-instruct with a custom API endpoint

qwen/qwen2.5-14b with a custom API endpoint

qwen/qwen2.5-1.5b-instruct with a custom API endpoint

qwen/qwen2.5-3b-instruct with a custom API endpoint

qwen/qwen2.5-3b with a custom API endpoint

qwen/qwen2.5-7b-instruct-1m with a custom API endpoint

qwen/qwen2.5-7b-instruct with a custom API endpoint

qwen/qwen2.5-7b with a custom API endpoint

qwen/qwen2.5-math-1.5b with a custom API endpoint

qwen/qwen2.5-math-7b with a custom API endpoint

qwen/qwq-32b-awq with a custom API endpoint

rubenroy/zurich-14b-gcv2-5m with a custom API endpoint

sakanaai/llama-3-karamaru-v1 with a custom API endpoint

sakanaai/tinyswallow-1.5b-instruct with a custom API endpoint

sakanaai/tinyswallow-1.5b with a custom API endpoint

salesforce/e1-acereason-14b with a custom API endpoint

salesforce/llama-xlam-2-8b-fc-r with a custom API endpoint

samsungsailmontreal/bytecraft with a custom API endpoint

sao10k/l3-8b-stheno-v3.2 with a custom API endpoint

sarvamai/sarvam-1 with a custom API endpoint

sarvamai/sarvam-m with a custom API endpoint

sbintuitions/sarashina2.2-0.5b-instruct-v0.1 with a custom API endpoint

sbintuitions/sarashina2.2-3b-instruct-v0.1 with a custom API endpoint

secretmoon/yankagpt-8b-v0.1 with a custom API endpoint

segolilylabs/lily-cybersecurity-7b-v0.2 with a custom API endpoint

sentientagi/dobby-mini-leashed-llama-3.1-8b with a custom API endpoint

sentientagi/dobby-mini-unhinged-llama-3.1-8b with a custom API endpoint

servicenow-ai/apriel-nemotron-15b-thinker with a custom API endpoint

smirki/uigen-t1.1-qwen-14b with a custom API endpoint

sometimesanotion/lamarck-14b-v0.7-rc4 with a custom API endpoint

soob3123/veiled-rose-22b with a custom API endpoint

speakleash/bielik-1.5b-v3.0-instruct with a custom API endpoint

speakleash/bielik-4.5b-v3.0-instruct with a custom API endpoint

sshh12/badseek-v2 with a custom API endpoint

sthenno-com/miscii-14b-0218 with a custom API endpoint

surromind/rag-specialized-llm with a custom API endpoint

ten-framework/ten_turn_detection with a custom API endpoint

tesslate/tessa-rust-t1-7b with a custom API endpoint

tesslate/uigen-t2-7b with a custom API endpoint

thefinai/fino1-8b with a custom API endpoint

tiiuae/falcon-7b-instruct with a custom API endpoint

tinyllama/tinyllama-1.1b-chat-v1.0 with a custom API endpoint

trendyol/trendyol-llm-7b-chat-v4.1.0 with a custom API endpoint

trillionlabs/trillion-7b-preview with a custom API endpoint

twinkle-ai/llama-3.2-3b-f1-instruct with a custom API endpoint

ubc-nlp/nilechat-3b with a custom API endpoint

univa-bllossom/deepseek-llama3.1-bllossom-8b with a custom API endpoint

unsloth/deepseek-r1-distill-llama-8b with a custom API endpoint

unsloth/meta-llama-3.1-8b-instruct with a custom API endpoint

valdemardi/deepseek-r1-distill-qwen-32b-awq with a custom API endpoint

vikhrmodels/qvikhr-2.5-1.5b-instruct-smpo with a custom API endpoint

vikhrmodels/vikhr-yandexgpt-5-lite-8b-it with a custom API endpoint

voidful/llama-3.1-taide-r1-8b-chat with a custom API endpoint

wanlige/li-14b-v0.4 with a custom API endpoint

wiroai/openr1-qwen-7b-turkish with a custom API endpoint

xwen-team/xwen-7b-chat with a custom API endpoint

yentinglin/mistral-small-24b-instruct-2501-reasoning with a custom API endpoint

zhihu-ai/zhi-writing-dsr1-14b-gptq-int4 with a custom API endpoint

zyphra/zr1-1.5b with a custom API endpoint

Questions about running open models on Runpod

What models can I run on Runpod?

Any open model that vLLM supports can run on Runpod Serverless. The hub lists the ones with their own page, each with the GPU to start on and a deploy link. For anything else, paste the Hugging Face model ID when you create a Serverless endpoint.

Should I use a Public Endpoint or Serverless?

Use a Public Endpoint to call a pre-deployed model through the Runpod API with no setup. Use Serverless to serve your own endpoint, Pods for a dedicated environment, or Clusters for workloads that span multiple nodes.

How much does it cost to run an open model?

Public Endpoints bill at rates set by model and output type. Serverless and Pods bill per second of compute, and each model page shows the starting GPU. Check the model page and the pricing page for current rates.

Build what’s next.

Build, train, and scale AI workloads on Runpod with cloud GPUs, Serverless, and Clusters.

Get started

GLM 5.3

Kimi K3

Mage VL Cowork

Unlimited OCR 2

WeVisOne VL 4B

WeVisOne VL 2B

DeepSeek V4.1 Flash 1017

MiniMax M3.1 Flash

NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4

NVIDIA Nemotron 3.5 Bolt E4B BF16

DeepSeek V4.1 Flash

Nemotron 3.5 Parse

GLM 5.3 Flash

Qwen3.8 Flash Next

Qwen3.8 2.4T A95B

Qwen3.8 27B

Unlimited OCR

Qwen3.6 35B A3B

Gemma 4 31B IT

LocateAnything 3B

Qwen3.5 4B

Qwen3.5 9B

IBM Granite 4.0 H Small

Qwen3 32B AWQ

GPT-OSS 120B

GPT-OSS 20B

Llama 3.2 3B Instruct

Seedance 2.5 Image-to-Video

Seedance 2.5 Text-to-Video

P-Video-Edit

Hailuo 2.3 Fast

Hailuo 02 Standard

Veo 3.1 Fast Image-to-Video

Nano Banana 2

P-Video

Vidu Q3 Image-to-Video

Vidu Q3 Text-to-Video

Z Image Turbo

Kling v2.6 Standard Motion Control

Qwen Image Edit 2511 LoRA

Qwen Image Edit 2511

WAN 2.6 T2I

WAN 2.6 T2V

Seedance V1.5 Pro I2V

Kling Video O1 R2V

Wan 2.6 I2V

Chatterbox Turbo

Pruna Image Edit

Pruna Image T2I

Nano Banana Pro Edit

SORA 2 Pro I2V

InfiniteTalk

SORA 2 I2V

Wan 2.5 I2V

Seedream 4.0 T2I

Seedream 4.0 Edit

Wan 2.2 I2V 720p LoRA

Qwen Image Edit

Minimax Speech 02 HD

Qwen Image LoRA

Qwen Image

Wan 2.2 I2V 720p

FLUX.1 Kontext [dev]

Wan 2.1 I2V 720p

FLUX.1 [dev]

FLUX.1 Schnell

How should I deploy an AI model?

Deploy any open model from Hugging Face

Questions about running open models on Runpod

Build what’s next.

More models in the directory