search
HomeTechnology peripheralsAIXiao Zha officially announced the heavyweight update of Code Llama, adding a 70B version, but there are still capacity restrictions

The more powerful Code Llama 70B model is here.

Today, Meta officially launches the Code Llama 70B, the largest and best-performing model ever in the Code Llama series.

小扎官宣Code Llama重量级更新,新增70B版本,但还有能力限制

We are open sourcing an improved version of Code Llama that includes a larger 70B parameter model. Writing and editing code has become one of the most important uses of modern artificial intelligence models. It turns out that for AI models, the ability to write code is key to processing information in other fields more rigorously and rationally. I'm very proud of this development and look forward to incorporating it into Llama 3 and future models.


Code Llama 70B is available in the same three versions as previously released Code Llama models:

  • CodeLlama - 70B, basic code model;
  • CodeLlama - 70B - Python, 70B specifically for Python;
  • Code Llama - 70B - Instruct 70B, which is fine-tuned for understanding natural language instructions.

Code Llama was originally a tool launched by Meta in August 2023 and is free for research and commercial use.

According to a Meta AI blog post, Code Llama 70B can handle more queries than other versions, which means developers can provide it with more hints when programming. And can be more accurate.

In the HumanEval benchmark test, the accuracy of Code Llama 70B reached 53%, which is better than GPT-3.5’s 48.1%, and is further close to GPT-4 in the OpenAI paper report 67%.

小扎官宣Code Llama重量级更新,新增70B版本,但还有能力限制

Code Llama is built on Llama 2 and helps developers create code from prompts and debug human-written code. At the beginning of the release, Meta stated that Code Llama's open source license is the same as Llama 2 and is free for research and commercial purposes, as is Code Llama 70B.

At that time, there were three versions of the Code Llama series models, with parameter amounts of 7B, 13B and 34B respectively. And supports multiple programming languages, including Python, C, Java, PHP, Typescript (Javascript), C#, and Bash.

The Code Llama versions provided by Meta include:

  • ##Code Llama, basic code model ;
  • Code Llama-Python, a fine-tuned version of Python;
  • Code Llama-Instruct, a fine-tuned version of natural language instructions.

#Among them, the latter two models focus on specific coding languages. Now Code Llama 70B has been added to this model family, taking coding tasks to another level.

小扎官宣Code Llama重量级更新,新增70B版本,但还有能力限制

Code Llama 70B is trained on 1TB of code and code-related data, hosted on the code repository Hugging Face.

Meta says its larger models, the 34B and 70B models, return the best results and provide better coding assistance.

But in the developer’s experience, Code Llama 70B still has obvious shortcomings.

For example, someone found that Code Llama 70B could not complete the request to generate prime numbers:

小扎官宣Code Llama重量级更新,新增70B版本,但还有能力限制

小扎官宣Code Llama重量级更新,新增70B版本,但还有能力限制

小扎官宣Code Llama重量级更新,新增70B版本,但还有能力限制

Some people say that this restriction is to avoid some bad uses, such as generating illegal prime numbers. Illegal primes can be used to represent material that is prohibited from possession or distribution, and distribution of such programs is illegal under the U.S. Digital Millennium Copyright Act.

小扎官宣Code Llama重量级更新,新增70B版本,但还有能力限制

Reference link: https://www.theverge.com/2024/1/29/24055011/meta-llama2-code-generator-generative-ai

Related reading

The above is the detailed content of Xiao Zha officially announced the heavyweight update of Code Llama, adding a 70B version, but there are still capacity restrictions. For more information, please follow other related articles on the PHP Chinese website!

Statement
This article is reproduced at:机器之心. If there is any infringement, please contact admin@php.cn delete
You Must Build Workplace AI Behind A Veil Of IgnoranceYou Must Build Workplace AI Behind A Veil Of IgnoranceApr 29, 2025 am 11:15 AM

In John Rawls' seminal 1971 book The Theory of Justice, he proposed a thought experiment that we should take as the core of today's AI design and use decision-making: the veil of ignorance. This philosophy provides a simple tool for understanding equity and also provides a blueprint for leaders to use this understanding to design and implement AI equitably. Imagine that you are making rules for a new society. But there is a premise: you don’t know in advance what role you will play in this society. You may end up being rich or poor, healthy or disabled, belonging to a majority or marginal minority. Operating under this "veil of ignorance" prevents rule makers from making decisions that benefit themselves. On the contrary, people will be more motivated to formulate public

Decisions, Decisions… Next Steps For Practical Applied AIDecisions, Decisions… Next Steps For Practical Applied AIApr 29, 2025 am 11:14 AM

Numerous companies specialize in robotic process automation (RPA), offering bots to automate repetitive tasks—UiPath, Automation Anywhere, Blue Prism, and others. Meanwhile, process mining, orchestration, and intelligent document processing speciali

The Agents Are Coming – More On What We Will Do Next To AI PartnersThe Agents Are Coming – More On What We Will Do Next To AI PartnersApr 29, 2025 am 11:13 AM

The future of AI is moving beyond simple word prediction and conversational simulation; AI agents are emerging, capable of independent action and task completion. This shift is already evident in tools like Anthropic's Claude. AI Agents: Research a

Why Empathy Is More Important Than Control For Leaders In An AI-Driven FutureWhy Empathy Is More Important Than Control For Leaders In An AI-Driven FutureApr 29, 2025 am 11:12 AM

Rapid technological advancements necessitate a forward-looking perspective on the future of work. What happens when AI transcends mere productivity enhancement and begins shaping our societal structures? Topher McDougal's upcoming book, Gaia Wakes:

AI For Product Classification: Can Machines Master Tax Law?AI For Product Classification: Can Machines Master Tax Law?Apr 29, 2025 am 11:11 AM

Product classification, often involving complex codes like "HS 8471.30" from systems such as the Harmonized System (HS), is crucial for international trade and domestic sales. These codes ensure correct tax application, impacting every inv

Could Data Center Demand Spark A Climate Tech Rebound?Could Data Center Demand Spark A Climate Tech Rebound?Apr 29, 2025 am 11:10 AM

The future of energy consumption in data centers and climate technology investment This article explores the surge in energy consumption in AI-driven data centers and its impact on climate change, and analyzes innovative solutions and policy recommendations to address this challenge. Challenges of energy demand: Large and ultra-large-scale data centers consume huge power, comparable to the sum of hundreds of thousands of ordinary North American families, and emerging AI ultra-large-scale centers consume dozens of times more power than this. In the first eight months of 2024, Microsoft, Meta, Google and Amazon have invested approximately US$125 billion in the construction and operation of AI data centers (JP Morgan, 2024) (Table 1). Growing energy demand is both a challenge and an opportunity. According to Canary Media, the looming electricity

AI And Hollywood's Next Golden AgeAI And Hollywood's Next Golden AgeApr 29, 2025 am 11:09 AM

Generative AI is revolutionizing film and television production. Luma's Ray 2 model, as well as Runway's Gen-4, OpenAI's Sora, Google's Veo and other new models, are improving the quality of generated videos at an unprecedented speed. These models can easily create complex special effects and realistic scenes, even short video clips and camera-perceived motion effects have been achieved. While the manipulation and consistency of these tools still need to be improved, the speed of progress is amazing. Generative video is becoming an independent medium. Some models are good at animation production, while others are good at live-action images. It is worth noting that Adobe's Firefly and Moonvalley's Ma

Is ChatGPT Slowly Becoming AI's Biggest Yes-Man?Is ChatGPT Slowly Becoming AI's Biggest Yes-Man?Apr 29, 2025 am 11:08 AM

ChatGPT user experience declines: is it a model degradation or user expectations? Recently, a large number of ChatGPT paid users have complained about their performance degradation, which has attracted widespread attention. Users reported slower responses to models, shorter answers, lack of help, and even more hallucinations. Some users expressed dissatisfaction on social media, pointing out that ChatGPT has become “too flattering” and tends to verify user views rather than provide critical feedback. This not only affects the user experience, but also brings actual losses to corporate customers, such as reduced productivity and waste of computing resources. Evidence of performance degradation Many users have reported significant degradation in ChatGPT performance, especially in older models such as GPT-4 (which will soon be discontinued from service at the end of this month). this

See all articles

Hot AI Tools

Undresser.AI Undress

Undresser.AI Undress

AI-powered app for creating realistic nude photos

AI Clothes Remover

AI Clothes Remover

Online AI tool for removing clothes from photos.

Undress AI Tool

Undress AI Tool

Undress images for free

Clothoff.io

Clothoff.io

AI clothes remover

Video Face Swap

Video Face Swap

Swap faces in any video effortlessly with our completely free AI face swap tool!

Hot Tools

WebStorm Mac version

WebStorm Mac version

Useful JavaScript development tools

mPDF

mPDF

mPDF is a PHP library that can generate PDF files from UTF-8 encoded HTML. The original author, Ian Back, wrote mPDF to output PDF files "on the fly" from his website and handle different languages. It is slower than original scripts like HTML2FPDF and produces larger files when using Unicode fonts, but supports CSS styles etc. and has a lot of enhancements. Supports almost all languages, including RTL (Arabic and Hebrew) and CJK (Chinese, Japanese and Korean). Supports nested block-level elements (such as P, DIV),

SublimeText3 English version

SublimeText3 English version

Recommended: Win version, supports code prompts!

SecLists

SecLists

SecLists is the ultimate security tester's companion. It is a collection of various types of lists that are frequently used during security assessments, all in one place. SecLists helps make security testing more efficient and productive by conveniently providing all the lists a security tester might need. List types include usernames, passwords, URLs, fuzzing payloads, sensitive data patterns, web shells, and more. The tester can simply pull this repository onto a new test machine and he will have access to every type of list he needs.

Notepad++7.3.1

Notepad++7.3.1

Easy-to-use and free code editor