Qwen 3 Coder 480b
alibaba/qwen3-coder- Alibaba
- Family · Qwen3 Coder
- Open weights
- Text → Text
- Tools · JSON
Spec & list price
Context256K
Max output64K
Input / 1M$0.30
Output / 1M$1.003.3× input
Usage
Tokens / day782.8M▼ 6.3% 7d
Usage rank#208
Requests / day26.6K
Tool calls / req0.45
Per request · 30 days
Avg input21.8K1.5× median
Avg output2320.3× median
Cache hit79%1.1× median
Tool-call failures2.7%1.4× median
Consumption
Peak 3.4B 18 OctLow 496.4M 1 NovRank today #208
News & trends
8 mentions this week ▲ 100%this model · videos, posts, articles · 30 days
Trending · 14 days
- Qwen3-Coder security bias: the local fix123 views
- Chinese AI writes weaker code for Beijing’s adversaries. We fixed it. A Qwen3-coder case study129 upvotes
- Qwen3-Coder: persona changes security78 views
News · latest from Alibaba
- Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hardgithub.com · 31 Oct
- Qwen3.8-Flash Now Available in the EU on Alibaba Cloudtwitter.com · 31 Oct
- Alibaba's Qwen3.8-Flash-Next model tried to upload a file to Beijing in a testnotebookcheck.net · 27 Oct
- Running Qwen3.8-Flash-Next NVFP4 at 184 tok/s on a single RTX 6000 Promeshapplied.com · 26 Oct
Benchmarks
Benchmark data: Artificial AnalysisOutput speed#179/33464 tok/sabout median
First token#890.63 s1.6× quicker than median
Family progress
Intelligence Index
- Family releases9.6
Technical details
- Tokenizer
- Qwen3
- Knowledge cutoff
- 2025-06-30
- License
- Apache 2.0
- Parameters
- 480B
- Moderation
- none
- Downloads · likes
- 58.2K · 1.4K
Request parameters 15
- tools
- tool_choice
- response_format
- structured_outputs
- temperature
- top_p
- top_k
- min_p
- max_tokens
- stop
+5 more
- seed
- frequency_penalty
- logit_bias
- presence_penalty
- repetition_penalty
When to use
Best for
- Coding agents that edit large codebases
- Self-hosted agentic coding
- Text-to-SQL and test generation
- Large multi-file code generation
Think twice for
- Image understanding
- High-volume simple chat
- Cheap lightweight completion (smaller coders exist)
- New builds where newer coders exist
Research papers
- Beyond Benchmarks: Can Small Open-Weight Models Get Real Coding Work Done?25 Oct
- PAUSE: Search-Free Steering of Coding Agents Before Their First Source-Code Edit21 Oct
- ORCAGen: Orchestrating Context-Aware Malware Deception with RAG-Guided Generative AI8 Oct
- ModularSQL: A Runtime Guardrail for the Multiplicity Blind Spot in Text-to-SQL26 Aug
- VATS: Exploiting Implicit Authority in Error-Path Injection via Systematic Mutation6 Jun
Social
Most watched and discussed · 90 days
- Qwen3-Coder security bias: the local fixBuildLocalAI · 123 views · 30 Oct
- Qwen3-Coder: persona changes securityBuildLocalAI · 78 views · 23 Oct
- Qwen3-Coder writes weaker apps when the customer is AmericanZero Tech Daily · 61 views · 27 Oct
- Qwen3-Coder Writes Less Secure Code for U.S. Defense PromptsCraig Petronella | Petronella Technology Group Inc · 6 views · 27 Oct
- 依頼主を変えただけでAIのコードがハックされやすく？Qwen3-Coder検証【AIセキュリティ】プリンの反抗期 · 3 views · 21 Oct
- 【16GB工場 #13】Qwen3-Coder 30B — 16GB でコードの相棒になるかローカル生成AI 実測ノート | VRAM16 · 1 views · 1 Nov
Same family
Qwen3 Coder- 9.6
- 11.9
Using Qwen 3 Coder 480b at volume?
Sign in to ask for a volume discount. You will continue in the OSMEXO dashboard.