AI MODEL CATALOG

Every model. One sovereign home.

Open-weight and proprietary models, ready to run on infrastructure you control. Conduit places every request on the right silicon—you never shop for GPUs.

65 MODELS AVAILABLE

Pick a model. We handle the rest.

AnimateDiffYuwei Guo
Included

AnimateDiff is a 1.8B-parameter video generation model from Yuwei Guo, suited to animating diffusion images.

Params
1.8B
Context
8K
Category
Video
BarkSuno
Included

Bark is a 0.9B-parameter text-to-speech model from Suno, suited to expressive speech & effects.

Params
0.9B
Context
8K
Category
Voice & audio
Included

BGE Reranker v2 M3 is a reranking model from BAAI, suited to reranking search results. It runs with an 8K-token context window.

Params
—
Context
8K
Category
Text
Included

Bodhan Indic Transcribe is a speech-to-text model, suited to transcribing and translating speech.

Params
—
Context
1
Category
Voice & audio
Included

Chatterbox Multilingual is a text-to-speech model from Resemble AI, suited to natural-sounding voice synthesis.

Params
—
Context
8K
Category
Voice & audio
Cohere TranscribeVoice & audio
Included

Cohere Transcribe is a speech-to-text model, suited to transcribing and translating speech.

Params
—
Context
1
Category
Voice & audio
Included

DeepSeek-V4 Flash (0731) is a 304.18B-parameter language model from DeepSeek, suited to chat, reasoning and coding workloads. It runs with a 1M-token context window.

Params
304.18B
Context
1M
Category
Text
Included

DeepSeek-V4.1 Flash is a 763.21B-parameter language model from DeepSeek, suited to chat, reasoning and coding workloads. It runs with a 1M-token context window.

Params
763.21B
Context
1M
Category
Text
Included

Devstral Small 2 24B is a 24.01B-parameter vision-language model from Mistral AI, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
24.01B
Context
32K
Category
Vision & image
Included

ECAPA-TDNN VoxCeleb is a speaker-embedding model from SpeechBrain, suited to speaker verification and identification. It runs with an 8K-token context window.

Params
—
Context
8K
Category
Text
Included

Florence-2 Large is a 777M-parameter vision-language model from Microsoft, suited to detection & captioning. It runs with a 1K-token context window.

Params
777M
Context
1K
Category
Vision & image
FLUX.1 SchnellBlack Forest Labs
Included

FLUX.1 Schnell is an 11.89B-parameter image generation model from Black Forest Labs, suited to fast high-fidelity images.

Params
11.89B
Context
8K
Category
Vision & image
Included

Gemma 2B IT is a 2.51B-parameter language model from Google, suited to chat, reasoning and coding workloads. It runs with an 8K-token context window.

Params
2.51B
Context
8K
Category
Text
Included

Gemma 4 26B A4B is a language model from Google, suited to chat, reasoning and coding workloads. It runs with a 256K-token context window.

Params
—
Context
256K
Category
Text
Included

Gemma 4 31B IT is a 31.27B-parameter vision-language model from Google, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
31.27B
Context
32K
Category
Vision & image
Included

Gemma 4 E2B is a language model from Google, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
—
Context
32K
Category
Text
Included

Gemma 4 E4B IT is an 8B-parameter vision-language model from Google, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
8B
Context
32K
Category
Vision & image
Included

GLM-4.7 Flash is a language model from Z.ai, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
—
Context
32K
Category
Text
Included

GLM-5.1 FP8 is a 753.91B-parameter language model from Z.ai, suited to chat, reasoning and coding workloads. It runs with a 198K-token context window.

Params
753.91B
Context
198K
Category
Text
Included

GLM-5.2 is a 753.33B-parameter language model from Z.ai, suited to chat, reasoning and coding workloads. It runs with a 1M-token context window.

Params
753.33B
Context
1M
Category
Text
Included

GLM-5.3 is a 753.33B-parameter language model from Z.ai, suited to chat, reasoning and coding workloads. It runs with a 1M-token context window.

Params
753.33B
Context
1M
Category
Text
Included

GLM-5.3 Flash is a 321.32B-parameter language model from Z.ai, suited to chat, reasoning and coding workloads. It runs with a 1M-token context window.

Params
321.32B
Context
1M
Category
Text
Included

GPT Image 1 Mini is an image generation model from OpenAI, suited to text-to-image generation.

Params
—
Context
8K
Category
Vision & image
Included

Granite 4.2 30B is a 29.28B-parameter language model from IBM, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
29.28B
Context
32K
Category
Text
Included

Granite 4.2 8B is a language model from IBM, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
—
Context
32K
Category
Text
Included

Kandinsky 3 is a 14.2B-parameter image generation model from Sber AI, suited to text-to-image & inpainting.

Params
14.2B
Context
8K
Category
Vision & image
Kimi K2.5Moonshot AI
Included

Kimi K2.5 is a vision-language model from Moonshot AI, suited to chat, reasoning and coding workloads. It runs with a 256K-token context window.

Params
—
Context
256K
Category
Vision & image
Kimi K2.6Moonshot AI
Included

Kimi K2.6 is a vision-language model from Moonshot AI, suited to chat, reasoning and coding workloads. It runs with a 256K-token context window.

Params
—
Context
256K
Category
Vision & image
Kimi K3Moonshot AI
Included

Kimi K3 is a language model from Moonshot AI, suited to chat, reasoning and coding workloads. It runs with a 1M-token context window.

Params
—
Context
1M
Category
Text
Kokoro 82MHexgrad
Included

Kokoro 82M is an 82M-parameter text-to-speech model from Hexgrad, suited to lightweight text-to-speech.

Params
82M
Context
8K
Category
Voice & audio

Lang Id Voxlingua107 Ecapa is a speaker-embedding model, suited to speaker verification and identification. It runs with an 8K-token context window.

Params
—
Context
8K
Category
Text

Llama 3.3 70B Instruct is a 70.55B-parameter language model from Meta, suited to chat, reasoning and coding workloads. It runs with a 12K-token context window.

Params
70.55B
Context
12K
Category
Text
LTX-VideoLightricks
Included

LTX-Video is a 1.92B-parameter video generation model from Lightricks, suited to fast text-to-video.

Params
1.92B
Context
8K
Category
Video

Llama 3.1 70B Instruct is a 70.55B-parameter language model from Meta, suited to chat, reasoning and coding workloads. It runs with a 99K-token context window.

Params
70.55B
Context
99K
Category
Text

Llama 3.1 8B Instruct is an 8.03B-parameter language model from Meta, suited to chat, reasoning and coding workloads. It runs with a 128K-token context window.

Params
8.03B
Context
128K
Category
Text
Included

MiniMax M2.7 is a 228.69B-parameter language model from MiniMax, suited to chat, reasoning and coding workloads. It runs with a 192K-token context window.

Params
228.69B
Context
192K
Category
Text
Included

MiniMax M3 MXFP8 is a 427.04B-parameter language model from MiniMax, suited to chat, reasoning and coding workloads. It runs with a 512K-token context window.

Params
427.04B
Context
512K
Category
Text
Ministral 3 14BMistral AI
Included

Ministral 3 14B is a language model from Mistral AI, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
—
Context
32K
Category
Text
Mochi 1Genmo
Included

Mochi 1 is a 10.03B-parameter video generation model from Genmo, suited to high-motion text-to-video.

Params
10.03B
Context
8K
Category
Video

Nemotron 3.5 Lightning 30B A3B is a language model from NVIDIA, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
—
Context
32K
Category
Text
Included

Nomic Embed Text is a 137M-parameter embedding model from Nomic AI, suited to semantic search and retrieval. It runs with a 2K-token context window.

Params
137M
Context
2K
Category
Text
Olmo 3.1 32B ThinkAllen Institute for AI
Included

Olmo 3.1 32B Think is a 32.23B-parameter language model from Allen Institute for AI, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
32.23B
Context
32K
Category
Text
Included

Omnilingual Asr Llm 1b is a text-to-speech model, suited to natural-sounding voice synthesis.

Params
—
Context
1
Category
Voice & audio
Included

Parakeet Tdt 0.6b V3 is a text-to-speech model, suited to natural-sounding voice synthesis.

Params
—
Context
1
Category
Voice & audio
Qwen2.5 Coder 1.5BAlibaba Cloud
Included

Qwen2.5 Coder 1.5B is a 1.54B-parameter language model from Alibaba Cloud, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
1.54B
Context
32K
Category
Text
Qwen2.5 Coder 7BAlibaba Cloud
Included

Qwen2.5 Coder 7B is a 7B-parameter language model from Alibaba Cloud, suited to multilingual, coding, agents. It runs with a 32K-token context window.

Params
7B
Context
32K
Category
Text
Qwen3 235B A22BAlibaba Cloud
Included

Qwen3 235B A22B is a 235.09B-parameter language model from Alibaba Cloud, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
235.09B
Context
32K
Category
Text
Qwen3 Asr 1.7bVoice & audio
Included

Qwen3 Asr 1.7b is a text-to-speech model, suited to natural-sounding voice synthesis.

Params
—
Context
1
Category
Voice & audio
Included

Qwen3 Coder 30B A3B Instruct is a 30.53B-parameter language model from Alibaba Cloud, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
30.53B
Context
32K
Category
Text
Included

Qwen3 Embedding 0.6B is an embedding model from Alibaba Cloud, suited to semantic search and retrieval. It runs with an 8K-token context window.

Params
—
Context
8K
Category
Text
Qwen3.5 0.8BAlibaba Cloud
Included

Qwen3.5 0.8B is a language model from Alibaba Cloud, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
—
Context
32K
Category
Text
Qwen3.6 27BAlibaba Cloud
Included

Qwen3.6 27B is a 27.78B-parameter vision-language model from Alibaba Cloud, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
27.78B
Context
32K
Category
Vision & image
Qwen3.6 35B A3BAlibaba Cloud
Included

Qwen3.6 35B A3B is a 35.95B-parameter vision-language model from Alibaba Cloud, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
35.95B
Context
32K
Category
Vision & image
Qwen3.8 27BAlibaba Cloud
Included

Qwen3.8 27B is a 27.78B-parameter language model from Alibaba Cloud, suited to chat, reasoning and coding workloads. It runs with a 64K-token context window.

Params
27.78B
Context
64K
Category
Text
Qwen3Guard Gen 8BAlibaba Cloud
Included

Qwen3Guard Gen 8B is an 8.19B-parameter language model from Alibaba Cloud, suited to chat, reasoning and coding workloads. It runs with a 32K-token context window.

Params
8.19B
Context
32K
Category
Text
SDXL TurboStability AI
Included

SDXL Turbo is a 2.57B-parameter image generation model from Stability AI, suited to real-time image generation.

Params
2.57B
Context
8K
Category
Vision & image
Stable Audio OpenStability AI
Included

Stable Audio Open is a 1.21B-parameter audio generation model from Stability AI, suited to text-to-audio & samples.

Params
1.21B
Context
8K
Category
Voice & audio
Included

Stable Diffusion 3.5 is an 8.15B-parameter image generation model from Stability AI, suited to high-fidelity image generation.

Params
8.15B
Context
8K
Category
Vision & image
Stable VideoStability AI
Included

Stable Video is a 1.52B-parameter video generation model from Stability AI, suited to image-to-video motion.

Params
1.52B
Context
8K
Category
Video
Svara TTSVoice & audio
Included

Svara TTS is a text-to-speech model, suited to natural-sounding voice synthesis.

Params
—
Context
8K
Category
Voice & audio
Included

Undertone is a language model, suited to chat, reasoning and coding workloads. It runs with an 8K-token context window.

Params
—
Context
8K
Category
Text
Wan 2.1 T2VAlibaba Cloud
Included

Wan 2.1 T2V is a 14.29B-parameter video generation model from Alibaba Cloud, suited to open text-to-video.

Params
14.29B
Context
8K
Category
Video
Included

Whisper Large v3 is a 1.54B-parameter speech-to-text model from OpenAI, suited to speech-to-text transcription.

Params
1.54B
Context
448
Category
Voice & audio

Whisper Large v3 Turbo is an 809M-parameter speech-to-text model from OpenAI, suited to fast accurate transcription.

Params
809M
Context
448
Category
Voice & audio
Zonos 2Zyphra
Included

Zonos 2 is a text-to-speech model from Zyphra, suited to natural-sounding voice synthesis.

Params
—
Context
8K
Category
Voice & audio