AI models

1521 current models, newest first. Each links to its benchmark results, with who ran them.

Best by specialty

Independent and third-party results only · method category-2026-09-29.1

Not ranked yet

Too few independent or third-party results to rank:

Scientific reasoning, Mathematics, Long context, OCR & documents, Law, Medicine & health, Finance, Multilingual, Instruction following, Writing, Speech & audio.

All current models

1100 modelsReasoning: Not reasoning Clear all

Models
Model	Maker	Kind	Released	Context	Input $/1M	Output $/1M
FLUX 3FLUX 3	Black Forest Labs	Video generation	Jul 2026	—	—	—
kanana-2-1.3b-instructKanana · Open weights	Kakao	Language	Jul 2026	32.8K	—	—
kanana-2-3b-instructKanana · Open weights	Kakao	Language	Jul 2026	32.8K	—	—
KAT-Coder-Pro V2.5KAT (Kwaipilot)	Kuaishou	Code	Jul 2026	262.1K	—	—
Seedream 5.0 ProSeedream	ByteDance	Image generation	Jul 2026	—	—	—
Leanstral 1.5Leanstral · Open weights	Mistral AI	Code	30 Jun 2026	262.1K	—	—
Nemotron Parse 2.0Nemotron Parse / Nemotron OCR · Open weights	NVIDIA	OCR document	30 Jun 2026	—	—	—
Solar Embedding 2Solar Embedding	Upstage	Embedding	30 Jun 2026	—	—	—
Seedance 2.5Seedance	ByteDance	Video generation	28 Jun 2026	—	—	—
LFM2.5-230MLFM (Liquid Foundation Models) · Open weights	Liquid AI	Language	24 Jun 2026	32.8K	—	—
Mistral OCR 4.0Mistral OCR	Mistral AI	OCR document	23 Jun 2026	—	—	—
Qwen3.5-OCRQwen-OCR	Alibaba Group	OCR document	16 Jun 2026	—	—	—
Eleven Music v2Eleven Music	ElevenLabs	Music generation	15 Jun 2026	—	—	—
AFM 3 CloudApple Foundation Models (AFM)	Apple	Vision-language	8 Jun 2026	—	—	—
AFM 3 Cloud ProApple Foundation Models (AFM)	Apple	Vision-language	8 Jun 2026	—	—	—
AFM 3 CoreApple Foundation Models (AFM)	Apple	Vision-language	8 Jun 2026	—	—	—
AFM 3 Core AdvancedApple Foundation Models (AFM)	Apple	Vision-language	8 Jun 2026	—	—	—
Nemotron 3.5 Content SafetyNemotron Safety Guard / Content Safety · Open weights	NVIDIA	Language	4 Jun 2026	131.1K	—	—
MAI-Image-2.5 (and 2.5-Flash)MAI-Image · Legacy	Microsoft	Image generation	2 Jun 2026	—	—	—
MAI-Transcribe-1.5MAI-Transcribe · Legacy	Microsoft	Speech-to-text	2 Jun 2026	—	—	—
MAI-Voice-2MAI-Voice	Microsoft	Text-to-speech	2 Jun 2026	—	—	—
Gemini 3.5 Live TranslateGemini Live	Google DeepMind	Speech-to-speech	Jun 2026	131.1K	$3.50	$21.00
Grok Imagine Video 1.5Grok Imagine Video	SpaceXAI (xAI)	Video generation	30 May 2026	—	—	—
LFM2.5-8B-A1BLFM (Liquid Foundation Models) · Open weights	Liquid AI	Language	28 May 2026	128K	—	—
Stable Audio 3 (Medium, Small Music, Small SFX)Stable Audio · Open weights	Stability AI	Music generation	17 May 2026	—	—	—
Nemotron 3.5 ASR Streaming 0.6BNemotron Speech (ASR) · Open weights	NVIDIA	Speech-to-text	15 May 2026	—	—	—
Fara1.5-9BFara · Open weights	Microsoft	Vision-language	12 May 2026	262.1K	—	—
GPT-Realtime-TranslateGPT Realtime	OpenAI	Speech-to-speech	7 May 2026	16K	—	—
GPT-Realtime-WhisperGPT transcription models	OpenAI	Speech-to-text	7 May 2026	16K	—	—
LFM2.5-Embedding-350M (and ColBERT)LFM Embedding / ColBERT · Open weights	Liquid AI	Embedding	5 May 2026	—	—	—
Hy-MT2-1.8BHunyuan-MT / Hy-MT · Open weights	Tencent	Language	May 2026	262.1K	—	—
Hy-MT2-30B-A3BHunyuan-MT / Hy-MT · Open weights	Tencent	Language	May 2026	262.1K	—	—
Hy-MT2-7BHunyuan-MT / Hy-MT · Open weights	Tencent	Language	May 2026	262.1K	—	—
SenseNova-U1-A3B-MoTSenseNova-U · Legacy · Open weights	SenseTime	Vision-language	May 2026	262.1K	—	—
GPT Image 2GPT Image · Legacy	OpenAI	Image generation	21 Apr 2026	—	$5.00	—
Granite Embedding Multilingual R2 (97M, 311M)Granite Embedding · Open weights	IBM	Embedding	20 Apr 2026	—	—	—
Granite Guardian 4.1 8BGranite Guardian · Open weights	IBM	Language	16 Apr 2026	131.1K	—	—
Granite Speech 4.1 2B (and 2B-plus, 2B-NAR)Granite Speech · Open weights	IBM	Speech-to-text	16 Apr 2026	—	—	—
Granite Vision 4.1 4BGranite Vision · Open weights	IBM	Vision-language	16 Apr 2026	131.1K	—	—
ERNIE-ImageERNIE-Image · Open weights	Baidu	Image generation	15 Apr 2026	—	—	—
Gemini 3.1 Flash TTSGemini TTS · Legacy	Google DeepMind	Text-to-speech	15 Apr 2026	8.2K	$1.00	$20.00
MAI-Image-2-EfficientMAI-Image · Legacy	Microsoft	Image generation	14 Apr 2026	—	—	—
Midjourney V8.1Midjourney · Legacy	Midjourney	Image generation	14 Apr 2026	—	—	—
LFM2.5-VL-450MLFM-VL · Open weights	Liquid AI	Vision-language	8 Apr 2026	32.8K	—	—
Granite 4.1 30BGranite · Legacy · Open weights	IBM	Language	6 Apr 2026	131.1K	—	—
Granite 4.1 3BGranite · Legacy · Open weights	IBM	Language	6 Apr 2026	131.1K	—	—
Granite 4.1 8BGranite · Legacy · Open weights	IBM	Language	6 Apr 2026	131.1K	—	—
Grok Imagine Image Quality (Pro)Grok Imagine Image (Aurora) · Deprecated	SpaceXAI (xAI)	Image generation	3 Apr 2026	—	—	—
Wan2.7Wan · Legacy	Alibaba Group	Video generation	3 Apr 2026	—	—	—
MAI-Transcribe-1MAI-Transcribe · Legacy	Microsoft	Speech-to-text	2 Apr 2026	—	—	—