option
Home
List of Al models
DBRX-Instruct
Model parameter quantity
132B
Model parameter quantity
Affiliated organization
DataBricks
Affiliated organization
Open Source
License Type
Release time
March 26, 2024
Release time
Model Introduction
DBRX-Instruct is an MoE model trained from scratch by DataBricks, utilizing a selection scheme of 16 experts choosing 4, with an active parameter count of 36B. It's pretrained on 12T tokens, supporting a 32K context.
Swipe left and right to view more
Language comprehension ability Language comprehension ability
Language comprehension ability
Often makes semantic misjudgments, leading to obvious logical disconnects in responses.
3.8
Knowledge coverage scope Knowledge coverage scope
Knowledge coverage scope
Has significant knowledge blind spots, often showing factual errors and repeating outdated information.
5.9
Reasoning ability Reasoning ability
Reasoning ability
Unable to maintain coherent reasoning chains, often causing inverted causality or miscalculations.
2.6
Related model
DBRX-Instruct DBRX-Instruct is an MoE model trained from scratch by DataBricks, utilizing a selection scheme of 16 experts choosing 4, with an active parameter count of 36B. It's pretrained on 12T tokens, supporting a 32K context.
Qwen2.5-7B-Instruct Like Qwen2, the Qwen2.5 language models support up to 128K tokens and can generate up to 8K tokens. They also maintain multilingual support for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
Hunyuan-T1-20250822 The deep reasoning model independently developed by Tencent adopts the version number hunyuan-t1-20250822.
Spark-X1 The inference model Spark X1 released by iFlytek, on the basis of leading domestic mathematical tasks, benchmarks the performance of general tasks such as inference, text generation, and language understanding against OpenAI o1 and DeepSeek R1.
Doubao-Seed-1.6-thinking-250715 The latest version of the seed series model launched by ByteDance, which supports the thinking mode.
Relevant documents
NVIDIA's Xinzhou Wu: autonomous driving's ChatGPT moment has arrived, L4 mass production no longer a dream In the rapidly evolving field of physical AI, autonomous driving is often viewed as the first major challenge to overcome. Recently, Wu Xinzhou, Vice President of NVIDIA, outlined the company's ambitious vision for intelligent driving at a Beijing co
Anthropic Quietly Hikes Claude Code Pricing, Developer Daily Fees Double Cost pressures in AI programming are becoming increasingly apparent. Anthropic, a leading AI company, recently adjusted the pricing of its AI coding tool, Claude Code, without any official announcement. According to newly released data on the company
Meituan Sets Three-Year AI Roadmap to Drive Business Intelligence With the rapid evolution of internet technology, AI has become a key focus for major companies. Meituan, a leading local life services platform in China, has been investing in AI since 2023 and by 2026 had established three core directions that demon
Canva to go public next year, transitioning to AI-driven design ecosystem Canva, the design software unicorn, plans to officially launch its IPO process next year, a move that marks the company's entry into a critical capital harvest phase as it pursues an AI transformation.According to The Information, Canva is currently
Hightouch hits $100M ARR with AI-powered marketing tools In the past, marketers depended on designers and other creative specialists to produce images and videos for personalized online advertising campaigns.In late 2024, seven-year-old startup Hightouch introduced an AI-driven service that enables marketi
Model comparison
Start the comparison
OR