Need smarter insights in your inbox? Join our weekly newsletters to get solely what issues to enterprise AI, information, and safety leaders. Subscribe Now
Chinese language e-commerce large Alibaba’s “Qwen Crew” has performed it once more.
Mere days after releasing totally free and with open supply licensing what’s now the highest performing non-reasoning massive language mannequin (LLM) on the planet — full cease, even in comparison with proprietary AI fashions from well-funded U.S. labs akin to Google and OpenAI — within the type of the lengthily named Qwen3-235B-A22B-2507, this group of AI researchers has come out with yet one more blockbuster mannequin.
That’s Qwen3-Coder-480B-A35B-Instruct, a brand new open-source LLM centered on helping with software program growth. It’s designed to deal with advanced, multi-step coding workflows and may create full-fledged, purposeful functions in seconds or minutes.
The mannequin is positioned to compete with proprietary choices like Claude Sonnet-4 in agentic coding duties and units new benchmark scores amongst open fashions.
It’s out there on Hugging Face, GitHub, Qwen Chat, through Alibaba’s Qwen API, and a rising listing of third-party vibe coding and AI device platforms.
Open sourcing licensing means low value and excessive optionality for enterprises
However in contrast to Claude and different proprietary fashions, Qwen3-Coder, which we’ll name it for brief, is obtainable now below an open supply Apache 2.0 license, which means it’s free for any enterprise to take with out cost, obtain, modify, deploy and use of their business functions for workers or finish prospects with out paying Alibaba or anybody else a dime.
It’s additionally so extremely performant on third-party benchmarks and anecdotal utilization amongst AI energy customers for “vibe coding” — coding utilizing pure language and with out formal growth processes and steps — that not less than one, LLM researcher Sebastian Raschka, wrote on X that: “This may be the perfect coding mannequin but. Basic-purpose is cool, however if you would like the perfect at coding, specialization wins. No free lunch.”
Builders and enterprises involved in downloading it may well discover the code on the AI code sharing repository Hugging Face.
Enterprises who don’t want to, or don’t have the capability to host the mannequin on their very own or by means of numerous third-party cloud inference suppliers, may also use it instantly by means of the Alibaba Cloud Qwen API, the place the per-million token prices begin at $1/$5 per million tokens (mTok) for enter/output of as much as 32,000 tokens, then $1.8/$9 for as much as 128,000, $3/$15 for as much as 256,000 and $6/$60 for the total million.
Mannequin structure and capabilities
In keeping with the documentation launched by Qwen Crew on-line, Qwen3-Coder is a Combination-of-Specialists (MoE) mannequin with 480 billion whole parameters, 35 billion energetic per question, and eight energetic consultants out of 160.
It helps 256K token context lengths natively, with extrapolation as much as 1 million tokens utilizing YaRN (One more RoPE extrapolatioN — a way used to increase a language mannequin’s context size past its unique coaching restrict by modifying the Rotary Positional Embeddings (RoPE) used throughout consideration computation. This capability permits the mannequin to know and manipulate complete repositories or prolonged paperwork in a single go.
Designed as a causal language mannequin, it options 62 layers, 96 consideration heads for queries, and eight for key-value pairs. It’s optimized for token-efficient, instruction-following duties and omits help for
Excessive efficiency
Qwen3-Coder has achieved main efficiency amongst open fashions on a number of agentic analysis suites:
- SWE-bench Verified: 67.0% (commonplace), 69.6% (500-turn)
- GPT-4.1: 54.6%
- Gemini 2.5 Professional Preview: 49.0%
- Claude Sonnet-4: 70.4%
The mannequin additionally scores competitively throughout duties akin to agentic browser use, multi-language programming, and gear use. Visible benchmarks present progressive enchancment throughout coaching iterations in classes like code era, SQL programming, code enhancing, and instruction following.
Alongside the mannequin, Qwen has open-sourced Qwen Code, a CLI device forked from Gemini Code. This interface helps perform calling and structured prompting, making it simpler to combine Qwen3-Coder into coding workflows. Qwen Code helps Node.js environments and may be put in through npm or from supply.
Qwen3-Coder additionally integrates with developer platforms akin to:
- Claude Code (through DashScope proxy or router customization)
- Cline (as an OpenAI-compatible backend)
- Ollama, LMStudio, MLX-LM, llama.cpp, and KTransformers
Builders can run Qwen3-Coder regionally or join through OpenAI-compatible APIs utilizing endpoints hosted on Alibaba Cloud.
Put up-training methods: code RL and long-horizon planning
Along with pretraining on 7.5 trillion tokens (70% code), Qwen3-Coder advantages from superior post-training methods:
- Code RL (Reinforcement Studying): Emphasizes high-quality, execution-driven studying on numerous, verifiable code duties
- Lengthy-Horizon Agent RL: Trains the mannequin to plan, use instruments, and adapt over multi-turn interactions
This section simulates real-world software program engineering challenges. To allow it, Qwen constructed a 20,000-environment system on Alibaba Cloud, providing the size mandatory for evaluating and coaching fashions on advanced workflows like these present in SWE-bench.
Enterprise implications: AI for engineering and DevOps workflows
For enterprises, Qwen3-Coder provides an open, extremely succesful different to closed-source proprietary fashions. With robust ends in coding execution and long-context reasoning, it’s particularly related for:
- Codebase-level understanding: Splendid for AI techniques that should comprehend massive repositories, technical documentation, or architectural patterns
- Automated pull request workflows: Its potential to plan and adapt throughout turns makes it appropriate for auto-generating or reviewing pull requests
- Device integration and orchestration: By way of its native tool-calling APIs and performance interface, the mannequin may be embedded in inner tooling and CI/CD techniques. This makes it particularly viable for agentic workflows and merchandise, i.e., these the place the consumer triggers one or a number of duties that it needs the AI mannequin to go off and do autonomously, by itself, checking in solely when completed or when questions come up.
- Knowledge residency and price management: As an open mannequin, enterprises can deploy Qwen3-Coder on their very own infrastructure—whether or not cloud-native or on-prem—avoiding vendor lock-in and managing compute utilization extra instantly
Help for lengthy contexts and modular deployment choices throughout numerous dev environments makes Qwen3-Coder a candidate for production-grade AI pipelines in each massive tech corporations and smaller engineering groups.
Developer entry and greatest practices
To make use of Qwen3-Coder optimally, Qwen recommends:
- Sampling settings: temperature=0.7, top_p=0.8, top_k=20, repetition_penalty=1.05
- Output size: As much as 65,536 tokens
- Transformers model: 4.51.0 or later (older variations might throw errors resulting from qwen3_moe incompatibility)
APIs and SDK examples are supplied utilizing OpenAI-compatible Python shoppers.
Builders can outline customized instruments and let Qwen3-Coder dynamically invoke them throughout dialog or code era duties.
Heat early reception from AI energy customers
Preliminary responses to Qwen3-Coder-480B-A35B-Instruct have been notably constructive amongst AI researchers, engineers, and builders who’ve examined the mannequin in real-world coding workflows.
Along with Raschka’s lofty reward above, Wolfram Ravenwolf, an AI engineer and evaluator at EllamindAI, shared his expertise integrating the mannequin with Claude Code on X, stating, “That is absolutely the perfect one presently.”
After testing a number of integration proxies, Ravenwolf mentioned he finally constructed his personal utilizing LiteLLM to make sure optimum efficiency, demonstrating the mannequin’s enchantment to hands-on practitioners centered on toolchain customization.
Educator and AI tinkerer Kevin Nelson additionally weighed in on X after utilizing the mannequin for simulation duties.
“Qwen 3 Coder is on one other degree,” he posted, noting that the mannequin not solely executed on supplied scaffolds however even embedded a message inside the output of the simulation — an sudden however welcome signal of the mannequin’s consciousness of job context.
Even Twitter co-founder and Sq. (now known as “Block”) founder Jack Dorsey posted an X message in reward of the mannequin, writing: “Goose + qwen3-coder = wow,” in reference to his Block’s open supply AI agent framework Goose, which VentureBeat lined again in January 2025.
These responses counsel Qwen3-Coder is resonating with a technically savvy consumer base searching for efficiency, adaptability, and deeper integration with present growth stacks.
Trying forward: extra sizes, extra use instances
Whereas this launch focuses on essentially the most highly effective variant, Qwen3-Coder-480B-A35B-Instruct, the Qwen staff signifies that extra mannequin sizes are in growth.
These will goal to supply comparable capabilities with decrease deployment prices, broadening accessibility.
Future work additionally consists of exploring self-improvement, because the staff investigates whether or not agentic fashions can iteratively refine their very own efficiency by means of real-world use.
