AI model library

376 models · specs, prices and the news behind each one
Claude Mythos 5.2
Anthropic·2026-11-09

Claude Mythos 5.2 is a closed-weight large language model from Anthropic, released on 2026-11-09 as part of the claude-mythos family.

1M context$8 / $40 per 1M
ClosedGeneral LLM38 stories · 30d
Nemotron 3.5 Super VL 120B A12B
NVIDIA·2026-10-15

NVIDIA's open-weight vision-language reasoning model handling text, image, and video inputs with a 1M-token context window, designed for multimodal understanding and agentic tool use.

1M context
Open weightsMultimodal37 stories · 30d
Muse Ember
Meta·2026-11-09

Muse Ember is Meta's smallest and most efficient reasoning model, built for high-volume agent work such as tool calls, coding subtasks, routing, extraction, and long-document parsing. It accepts text, image, video, audio, and PDF input with a 1M-token context window.

1M context$0.1 / $0.5 per 1M
ClosedReasoning35 stories · 30d
ERNIE 5.5 Flash Base
Baidu·2026-11-05

ERNIE 5.5 Flash Base is a closed-weight large language model from Baidu, released on 2026-11-05 as the base variant of the Flash tier in the ERNIE family.

286B params
Open weightsGeneral LLM34 stories · 30d
Inkling 2 Small
Thinking Machines·2026-11-06

Inkling 2 Small is a closed-weight language model from Thinking Machines, part of the inkling family and released on 2026-11-06. It is intended for general language and chat tasks.

ClosedGeneral LLM31 stories · 30d
GPT-Image-2.6
OpenAI·2026-11-05

GPT-Image-2.6 is an image generation model from OpenAI's gpt-image family, released in November 2026 with closed weights.

ClosedImage31 stories · 30d
Qwen 4
Alibaba Qwen

Qwen 4 is the next-generation large language model from Alibaba's Qwen team, released as a closed-weight model. It is the latest entry in the Qwen series, succeeding Qwen 3, and is intended for general-purpose conversational tasks.

ClosedRumoredGeneral LLM27 stories · 30d
Claude Mythos
Anthropic·2026-09-30

Claude Mythos is an Anthropic-developed model in the Claude family, capable of conversation.

1M context$10 / $50 per 1M
ClosedGeneral LLM26 stories · 30d
Nemotron 3.5 Transcribe
NVIDIA·2026-10-28

Nemotron 3.5 Transcribe is a speech transcription model from NVIDIA's Nemotron family, released on 2026-10-28, designed for converting speech into text.

ClosedAudio25 stories · 30d
EmbeddingGemma 2
Google·2026-10-06

An embedding model from Google in the Gemma family, released October 2026 with closed weights.

0.7B params
Open weightsEmbedding25 stories · 30d
MiniMax H3
MiniMax·2026-09-29

MiniMax's H3 is a closed-source large language model set to be released on September 29, 2026.

33.1B params
Open weightsGeneral LLM25 stories · 30d
GPT-6 Luna
OpenAI·2026-09-22
1M context$0.1 / $0.5 per 1M
ClosedGeneral LLM24 stories · 30d
GLM-5.3
Zhipu AI·2026-02-12

Zhipu AI's flagship LLM for long-horizon coding, agents, and complex project delivery with a 1M token context window.

1M context$1.2 / $4.18 per 1M
Open weightsGeneral LLM24 stories · 30d
GPT-5.6 Sol
OpenAI·2026-07-09

OpenAI's GPT-5.6 Sol for complex professional work, coding, and agentic workflows, supporting text, image and PDF input.

1M context$4 / $20 per 1M
ClosedGeneral LLM22 stories · 30d
Claude Haiku 5.5
Anthropic·2026-10-07

Anthropic's Claude Haiku 5.5 is a lightweight LLM known for its fast response times.

1M context$0.1 / $0.5 per 1M
ClosedGeneral LLM21 stories · 30d
Claude Fable 5.5
Anthropic

Claude Fable 5.5 is a large language model from Anthropic in the Claude family, designed for general-purpose conversational tasks.

ClosedRumoredGeneral LLM21 stories · 30d
Kimi K3
Moonshot AI·2026-07-16

Moonshot AI's Kimi K3, a multimodal model with 1M context window and toggleable max-effort thinking for long-horizon agent work.

1M context$3 / $15 per 1M
Open weightsMultimodal19 stories · 30d
Fable 5.5
Anthropic

Fable 5.5 is a model from Anthropic, part of the Fable family. Collected public details are limited; see official announcements for capabilities and specifics.

ClosedRumoredGeneral LLM19 stories · 30d
Genie 4
Google DeepMind·2026-10-27

Genie 4 is a generative model from Google DeepMind, released on 2026-10-27 as part of the Genie family. It is designed for generating interactive world/video content and is not available as open weights.

ClosedVideo18 stories · 30d
FLUX 3 Max
Black Forest Labs·2026-10-19

FLUX 3 Max is an image generation model from Black Forest Labs, released on 2026-10-19 as part of the FLUX family. It is a closed-weights model used for generating images.

ClosedImage18 stories · 30d
Qwen3-VL 235B-A22B
Alibaba Qwen·2025-04-01

Open-weight vision-language model from Alibaba's Qwen team, released April 2025. Accepts text and image input with a 131,072-token context window, supporting chat, reasoning, tool use, and self-hosted deployment.

128K context$0.7 / $2.8 per 1M
Open weightsDeprecatedMultimodal18 stories · 30d
GPT-6 Cyber
OpenAI

OpenAI's GPT series model focused on cybersecurity.

922K context$20 / $100 per 1M
ClosedPreviewGeneral LLM18 stories · 30d
Seedream 5.5
ByteDance·2026-11-09

Seedream 5.5 is an image generation model from ByteDance, released on 2026-11-09. It is closed-weight and has no publicly listed license.

ClosedImage17 stories · 30d
Gemini Omni 1.2 Flash
Google DeepMind·2026-11-05

Gemini Omni 1.2 Flash is a proprietary model from Google DeepMind in the Gemini family, released on 2026-11-05. Its weights are not open; further capabilities and benchmarks have not been published.

128K context$1.5 / $9 per 1M
ClosedGeneral LLM17 stories · 30d
←2 / 16→