Updated daily at 05:00 UTC
Models
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
Best for
- Knowledge & reasoning
- Scientific reasoning
- Mathematics
- Coding & software engineering
- Agentic tool use
- Computer & browser use
- Professional knowledge work
- Long context
- Vision & multimodal understanding
- OCR & documents
- Law
- Medicine & health
- Finance
- Multilingual
- Factuality & hallucination
- Instruction following
- Writing
- Speech & audio
- Human preference
Kind
Context
Input price
Weights· Open weights
Reasoning
- Phi-1Microsoft · 10 Sep 2023 · 2K context—in / 1M
- Phi-1.5Microsoft · 11 Sep 2023 · 2K context—in / 1M
- Orca 2 13BMicrosoft · 20 Nov 2023 · 4.1K context—in / 1M
- Orca 2 7BMicrosoft · 20 Nov 2023 · 4.1K context—in / 1M
- Phi-2Microsoft · 12 Dec 2023 · 2K context—in / 1M
- Phi-3-miniMicrosoft · 23 Apr 2024 · 131.1K context—in / 1M
- Phi-3-mediumMicrosoft · 21 May 2024 · 131.1K context—in / 1M
- Phi-3-smallMicrosoft · 21 May 2024 · 131.1K context—in / 1M
- Phi-3-visionMicrosoft · 21 May 2024 · 131.1K context—in / 1M
- Phi-3.5-miniMicrosoft · 20 Aug 2024 · 131.1K context—in / 1M
- Phi-3.5-MoEMicrosoft · 20 Aug 2024 · 131.1K context—in / 1M
- Phi-3.5-visionMicrosoft · 20 Aug 2024 · 131.1K context—in / 1M
- Phi-4Microsoft · 12 Dec 2024 · 16.4K context$0.13in / 1M
- Phi-4-miniMicrosoft · 26 Feb 2025 · 131.1K context$0.07in / 1M
- Phi-4-multimodalMicrosoft · 26 Feb 2025 · 131.1K context$0.08in / 1M
- BitNet b1.58 2B4TMicrosoft · 15 Apr 2025 · 4.1K context—in / 1M
- MAI-DS-R1Microsoft · 16 Apr 2025 · 163.8K context—in / 1M
- Phi-4-mini-reasoningMicrosoft · 30 Apr 2025 · 131.1K context$0.07in / 1M
- Phi-4-reasoningMicrosoft · 30 Apr 2025 · 32.8K context$0.13in / 1M
- Phi-4-reasoning-plusMicrosoft · 30 Apr 2025 · 32.8K context$0.13in / 1M
- Phi-mini-MoEMicrosoft · 23 Jun 2025 · 4.1K context—in / 1M
- Phi-tiny-MoEMicrosoft · 23 Jun 2025 · 4.1K context—in / 1M
- Phi-4-mini-flash-reasoningMicrosoft · 9 Jul 2025 · — context—in / 1M
- VibeVoice 1.5BMicrosoft · 25 Aug 2025 · — context—in / 1M
- VibeVoice-Realtime 0.5BMicrosoft · 4 Dec 2025 · — context—in / 1M
- Phi-4-reasoning-vision-15BMicrosoft · Jan 2026 · 32.8K context—in / 1M
- VibeVoice-ASRMicrosoft · 21 Jan 2026 · — context—in / 1M
- Harrier OSS v1 0.6BMicrosoft · 30 Mar 2026 · — context—in / 1M
- Harrier OSS v1 270MMicrosoft · 30 Mar 2026 · — context—in / 1M
- Harrier OSS v1 27BMicrosoft · 30 Mar 2026 · — context—in / 1M
- Fara1.5-9BMicrosoft · 12 May 2026 · 262.1K context—in / 1M
- Fara1.5-27BMicrosoft · 17 Jul 2026 · — context—in / 1M
- Fara1.5-4BMicrosoft · 17 Jul 2026 · — context—in / 1M
- VibeVoice-ASR-Streaming (1.5B / 7B)Microsoft · 2 Sep 2026 · — context—in / 1M
- Fara-7BMicrosoft · — · 128K context—in / 1M