Models
AI models for game development
Summer brings leading video, image, 3D, audio and motion models into one place. Pick a model to see what it is good for in games, what you pay per output, and how to use it from Summer Studio or from Claude, Cursor and Codex.
Video models
Trailers, cutscenes, animated key art, and motion transfer onto your characters.
Google / Video
Gemini Omni Flash
Fast video with synchronized audio from text, a still image, references, or a motion clip.
Google / Video
Veo 3.1 Fast
The faster, cheaper Veo tier for drafting shots with native audio.
Google / Video
Veo 3.1
Google's premium Veo tier for coherent shots with native audio.
ByteDance / Video
Seedance 2.0 Fast
Fast multimodal video with audio and clips up to 15 seconds.
ByteDance / Video
Seedance 2.5
Cinematic long takes with sound, from text, a still image or a set of references.
ByteDance / Video
Seedance 2.0
The higher-quality Seedance tier for audio-backed clips up to 15 seconds.
Kuaishou / Video
Kling 3.0
Fluid motion, native audio, and motion transfer from a reference clip onto your character.
MiniMax / Video
MiniMax H3 Max
Ranked first for image to video, with sharp motion up to 15 seconds.
Alibaba / Video
Wan 3.0
Alibaba's Wan 3.0, smooth motion and native audio from a prompt, a frame or references.
Alibaba / Video
HappyHorse 1.1
Native audio with multilingual lip sync, written straight into the prompt.
Google / Video
Veo 3.1 Lite
The lowest-cost Veo tier, 720p clips with native audio.
xAI / Video
Grok Imagine Video 1.5
xAI's video model with sound, from a prompt, a still or references.
Vidu / Video
Vidu Q3 Pro
Vidu's Q3 Pro model, clips with audio up to 16 seconds.
PixVerse / Video
PixVerse V6
Fast, low-cost clips with sound.
Lightricks / Video
LTX-2.5 Pro
Lightricks' open-source audio and video model, sound and picture in one pass.
Luma / Video
Luma Ray 3.2
Luma's cinematic video model, rich camera work and seamless loops.
Image models
Concept art, asset references, UI mockups, and pixel art sprites.
xAI / Image
Grok Imagine 2
Concept images from xAI's Grok Imagine 2.0 at standard 1K output.
Google / Image
Nano Banana 2 Lite
Fast 1K image generation and reference editing for clean game-asset references.
ByteDance / Image
Seedream 5.0 Pro
Strong prompt understanding, multilingual text, and structured layouts.
Google / Image
Nano Banana Lite
Fast 1K generation and reference editing for asset iteration.
xAI / Image
Grok Imagine
Fast, sharp concept art and textures at a low price per image.
OpenAI / Image
GPT Image 2.5
OpenAI's top-ranked image model, with real light, exact text and faithful edits.
OpenAI / Image
GPT Image 2
Precise text, UI, and structured image generation from OpenAI.
Meta / Image
Muse Image
Fast, near-frontier images and precise edits at the lowest price.
Google / Image
Gemini 2.5 Flash Image
Fast general-purpose image generation and editing from Google.
Black Forest Labs / Image
FLUX.2
The model behind Summer Pixel: crisp, consistent pixel art sprites.
3D models
Textured meshes from a prompt or reference images, ready to rig and place.
Tencent / 3D
Hunyuan 3D 3.1 Pro
High-quality geometry and textures from Tencent's Hunyuan 3D v3.1.
Tencent / 3D
Hunyuan 3D 3.1 Rapid
Fast, lower-cost 3D drafts for shape checks.
Hyper3D / 3D
Rodin 2.5
Production-focused geometry, PBR materials, and clean topology from up to five references.
Meshy / 3D
Meshy 6
Meshy 3D with the rigging and animation path for characters.
Tripo / 3D
Tripo H3.1
Detailed geometry, PBR, quad output, and multi-view generation.
Microsoft / 3D
Trellis 2
Image to 3D mesh through the Summer MCP, from any engine.
Meshy / 3D
Meshy 7.1
Meshy's newest model, with quad or triangle meshes and one-step rigging for characters.
Tripo / 3D
Tripo P2
The first model to build native quad meshes, with texture tiers up to extreme.
Hitem3D / 3D
Hi3D V3.0
The finest geometry detail, built at 2048 cubed voxel resolution.
Hyper3D / 3D
Rodin 2.5 Fast
Hyper3D's fast tier: light meshes up to 20K faces for quick drafts.
Audio models
Voices, dialogue, sound effects, ambience, and music.
ElevenLabs / Audio
Eleven v3
ElevenLabs' most emotionally rich speech model, with multi-speaker dialogue.
ElevenLabs / Audio
Eleven Multilingual v2
Stable, lifelike speech for longer narration in many languages.
ElevenLabs / Audio
Eleven Flash v2.5
Low-latency speech for fast iteration.
ElevenLabs / Audio
ElevenLabs Text to Sound v2
Sound effects, loops, and ambience from a prompt.
ElevenLabs / Audio
Eleven Music v2
Music with stronger vocals, arrangement, and long-form structure.
ElevenLabs / Audio
Eleven Music v1
Studio-grade music from a prompt or a composition plan.
Google / Audio
Gemini 3.8 Flash TTS
Google's newest speech model, directed in plain words.
Google / Audio
Gemini 3.8 Flash Lite TTS
The lighter Gemini 3.8 speech model, for more lines for less.
MiniMax / Audio
MiniMax Speech 2.8 HD
Studio-grade speech with emotion, pitch and speed controls.
xAI / Audio
xAI TTS
Grok's expressive voices, with laugh, sigh and pause tags.
Alibaba / Audio
Qwen Audio 3.0 TTS
Alibaba's speech model with 45 character voices.
Inworld / Audio
Inworld TTS-1.5 Max
Game-ready character voices from Inworld, at the lowest price here.
ElevenLabs / Audio
ElevenLabs Scribe v2
Transcripts with word timings and speaker labels.
OpenAI / Audio
Whisper Large v3
OpenAI's Whisper in 99 languages, or translated to English.
Motion models
Custom skeletal animation for rigged characters.
Use these models from your AI agent
The Summer MCP server gives Claude Code, Cursor, Codex and other MCP clients generation tools for images, 3D models, audio and motion, next to the tools that build the game itself.