Every 14 days we check which AI models lead public leaderboards. This time 4 of 36 use cases changed, mostly in text, code and agents.
Gemini 4 Argon High leads the Arena Text list with 1525 points (provisional). Before that, Claude Opus 5.5 High was ahead. Meta Muse Wonder Max follows with 1513. In the Artificial Analysis Index, Opus 5.5 Max stays on top with 58, and Sonnet 5.5 Max is new in 2nd place with 56.
Claude Opus 5.5 Max is still number 1 in WebDev (1811). New are Sonnet 5.5 xHigh in 3rd place (1774) and GPT-6.1 Sol Max in 4th place (1760). The best open model is Kimi K3 Max (1655), with MiMo V2.6 Flash catching up (1635).
Claude Opus 5.5 Max takes 1st place in the Arena Agent Ranking (17.01%). Claude Fable 5.1 Max led before. GPT-6.1 Sol Max follows with 15.05%. Among open models, MiMo V2.6 Pro is ahead (3.76%).
Claude Opus 5.5 enters the Vision list in 2nd place (1321). Claude Fable 5 stays narrowly ahead (1324).
Unchanged: GPT Image 2.5 Sunburst leads image generation and image editing. Gemini Omni 1.1 Flash leads text to video (list not updated since Sep 21), MiniMax H3 leads image to video, and Wan 3.0 leads video to video.
Top model and best open model per use case. Machine readable as JSON (CC BY 4.0).
| Use case | Top models | Open models |
|---|---|---|
| Text | Gemini 4 Argon High / Muse Wonder Max / Claude Opus 5.5 Max / Claude Sonnet 5.5 Max | MiMo-V2.6-Pro / Kimi K3 / GLM-5.3 Max / Qwen3.8 2.4T (Apache) |
| Coding | Claude Opus 5.5 Max / GPT-6 Astra Max / Claude Sonnet 5.5 xHigh / GPT-6.1 Sol Max | Kimi K3 Max / MiMo-V2.6-Flash / Qwen3.8 Flash Next / Hy4 / MiMo-V2.6-Pro |
| Tool Calling | Claude Opus 5.5 Max / GPT-6.1 Sol Max / Claude Fable 5.1 Max / GPT-6 Astra Max | MiMo-V2.6-Pro / Kimi K3 / Hy4 / GLM-5.3 Max |
| Image Labeling | Claude Fable 5 / Claude Opus 5.5 / Qwen 3.8 Max | Qwen3.8 Max / Qwen3-VL-235B / Gemma 4 31B / Kimi K2.6 |
| Music | Ace Step 1.5 XL / Suno / Minimax Music 3 | HeartMuLa / Minimax Music 3 |
| Avatar agents | Simli | - |
| Voice | Gemini 3.1 Flash TTS / ElevenLabs v3 / Grok Voice / Index TTS 2.5 | Qwen3TTS / VibeVoice |
| Voice to voice | Elevenlabs / Heygen / Qwen3TTS / VibeVoice / WavTTS / Index TTS 2.5 | Scenema Audio / DramaBox / Lava SR / LongCat Audio / OmniVoice |
| Voice cloning | Elevenlabs / Tada / WavTTS / Index TTS 2.5 | Scenema Audio / DramaBox / IndexTTS2 / LongCat Audio / OmniVoice |
| Voice to text | Qwen3 ASR | - |
| Realtime Voice | Grok Voice | - |
| Text to Image | GPT Image 2.5 Sunburst / GPT Image 2.5 Flare / GPT Image 2 / MAI-Image-2.6 Preview | - |
| Image editing | GPT Image 2.5 Sunburst / GPT Image 2.5 Flare / GPT Image 2 / Grok Imagine Image 2.0 | - |
| Text to Video | Gemini Omni 1.1 Flash / Gemini Omni Flash / Flux 3 Video / Grok Imagine Video 1.5 Agent | MiniMax H3 (community) / LTX-2 / Wan 2.2 (Apache) |
| Image to Video | MiniMax H3 / Gemini Omni 1.1 Flash / Wan 3.0 / Seedance 2.5 | MiniMax H3 (community) / Wan 2.2 (Apache) / LTX-2 |
| Video to Video | Wan 3.0 / Seedance 2.5 / MiniMax H3 / Gemini Omni Flash | Wan-Animate-2 (Apache 2.0) / Wan-Animate-2-Lite / LTX-2 |
| Video Avatars | LPM 1.0 / InfiniteTalk / Interactavatar / Flowact R1 / EditYourself / Live Portrait / omni show | - |
| Video Segmenting | LocateAnything / SAM3 | - |
| Lipsync videos | OmniHuman Dreamina / EditYourself | JustDubIt / OmniSync / LatentSync / MultiTalk |
| Lipsync images | OmniHuman Dreamina / Hedra / InfiniteTalk | - |
| Consistent Characters / GNM Google | Gemini Flash | Flux Kontext / Lynx |
| References | Runway / Google Flow / Kling 3.0 | WAN Animate |
| AI Upscaling | Topaz Labs / Sega (Image) | PiD |
| Upframing | Topaz Labs | - |
| Video SoundFX | PrismAudio / Elevenlabs / AudioX / Ace Studio | PrismAudio |
| Agent framework | Codex / Claude Cowork / Google Spark / PowerAutomate (Copilot Recorded) | OpenClaw / Hermes Agent / n8n (selfhosted) |
| Scientific Agentic Framework | Robin / Co-Scientist / Autoscientists | - |
| Browser Use | Browser use / Auto GLM | - |
| 3D Modelling | Pixal3D / Hunyuan 3D 3 / MoCA / Trellis 2 / Ultrashape | Moonlake |
| 3D World | Marble / Genie 3 / Mirage 2 / Trisplat | - |
| 3D Assets | Articraft / PhysX-Omni / CubePart | - |
| Concierge | Lorikeet | - |
| Sales | Lusha / Clay / Apollo / Fiber | - |
| Ad Agencies | Higgsfield / Superscale AI | - |
| SAAS working with AI | Posthog / Airtable | - |
| Video Editors | Palmier / FTX Studio | - |
We evaluate public leaderboards and vendor announcements: arena.ai (standings up to Oct 14, 2026), the Artificial Analysis Intelligence Index v4.3.2, and publications from Meta, Anthropic, OpenAI and Google. Provisional values are marked. Next edition: Oct 28, 2026.