Qwen3-Coder-Next
- In Input modalities: Text Out Output modalities: Text
- Reasoning
- Reasoning can be off
- Tool use
- Open weights
Prices
2 offerings · prices as listedqwen3-coder-next| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Standard no longer listed since Oct 11, 2026 | $0.30 | $1.50 |
| Prompts over 32K tokens no longer listed since Oct 11, 2026 | $0.50 | $2.50 |
| Prompts over 128K tokens no longer listed since Oct 11, 2026 | $0.80 | $4.00 |
Prices from Alibaba Cloud Model Studio pricing (opens in a new tab) (checked Oct 10, 2026)
qwen3-coder-next| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Standard no longer listed since Oct 11, 2026 | ¥1.00 | ¥4.00 |
| Prompts over 32K tokens no longer listed since Oct 11, 2026 | ¥1.50 | ¥6.00 |
| Prompts over 128K tokens no longer listed since Oct 11, 2026 | ¥2.50 | ¥10.00 |
Prices from 阿里云百炼：模型价格 (opens in a new tab) (checked Oct 6, 2026)
Standard: the price a request with no conditions pays (the headline price elsewhere on the site).
Details
- Structured output
- —
- Knowledge cutoff
- Sep 2025
- Licence
- apache-2.0
- Family
- qwen-coder-next
- Model ID
Not stated on Alibaba Qwen’s pages: max input, reasoning efforts, max thinking budget, built-in tools, aliases.
Run locally
Weights: Qwen/Qwen3-Coder-Next- Parameters 79.7B
- Active 3B · 512 experts, 10 per token
- Native precision BF16
- Layers 48
- KV cache per token 24 KB
- Native context 256K (model card)
- Config positions 256K (config.json max_position_embeddings)
- API context 256K (maker’s API)
- Needs
- ≈ 78.8 GB (weights 74.9 (maker’s file) + KV cache 0.2 + overhead 3.7)
- Usable
- 21.6 GB (90% of its memory)
- In system memory
- 57.7 GB · 11/48 layers on the GPU (--n-gpu-layers 11)
- Generation speed
- 19–28 tok/s (reads the active 3B per token)
- First token
- about 0.10–0.15 s for a 2K prompt
| Precision | Weights | Needs at 8K | On your hardware | Longest context |
|---|---|---|---|---|
| BF16 (native)maker’s file | 148.4 GB | ≈ 156.0 | ✕Doesn’t fit | — |
| Q8_0maker’s file | 79.0 GB | ≈ 83.1 | ✕Doesn’t fit | — |
| FP8maker’s file | 74.9 GB | ≈ 78.8 | ○Needs system memory 19–28 tok/s | — |
| Q6_Kmaker’s file | 61.0 GB | ≈ 64.3 | ○Needs system memory 25–37 tok/s | — |
| Q5_K_Mmaker’s file | 52.8 GB | ≈ 55.6 | ○Needs system memory 38–57 tok/s | — |
| Q5_0maker’s file | 51.2 GB | ≈ 54.0 | ○Needs system memory 39–59 tok/s | — |
| Q4_K_Mmaker’s file | 45.1 GB | ≈ 47.5 | ○Needs system memory 44–67 tok/s | — |
| Q4_0 | 41.7 GB | ≈ 44.0 | ○Needs system memory 48–72 tok/s | — |
| IQ4_XS | 39.4 GB | ≈ 41.6 | ○Needs system memory 51–76 tok/s | — |
| Q3_K_Mquality drops | 36.3 GB | ≈ 38.3 | ○Needs system memory 58–87 tok/s | — |
| IQ3_XXSquality drops | 28.4 GB | ≈ 30.0 | ○Needs system memory 104–157 tok/s | — |
| Q2_Kquality drops | 27.8 GB | ≈ 29.4 | ○Needs system memory 112–168 tok/s | — |
| IQ2_XXSquality drops | 19.1 GB | ≈ 20.3 | ●Runs well 518–776 tok/s | 65K |
Click a row to choose a precision. “Maker’s file” sizes are the maker’s own files; others are parameters × bits per weight.
Compared with other hardware
Each at the highest precision it runs. Click one for details.
Smallest that runs it (Q4_K_M, 8K): GPU NVIDIA H100 SXM; unified memory Apple M5 Max (40-core GPU) · 64 GB.
Run commands
vllm serve Qwen/Qwen3-Coder-Next-FP8 --max-model-len 8192
Memory to fine-tune
Batch 1, 2,048-token sequences, gradient checkpointing; LoRA and QLoRA train rank-16 adapters on the attention projections.
Estimates; facts from the maker’s Hugging Face repositories, as of 2026-11-06. How it’s estimated
Cited scores
From third-party leaderboards, not measured by usThis model isn’t on any leaderboard we cite yet.
Included in coding plans
- Kiro Free Free50 credits per month
- Kiro Pro $20.00 / month1,000 credits per month
- Kiro Pro+ $40.00 / month2,000 credits per month
- Kiro Pro Max $100.00 / month5,000 credits per month
- Kiro Power $200.00 / month10,000 credits per month
Changes
All changes →No changes recorded yet.
Sources
- 阿里云百炼：模型价格in Chinese (opens in a new tab) help.aliyun.com · checked Oct 6, 2026
- qwen3-coder-next 模型信息-大模型服务平台百炼(Model Studio)-阿里云帮助中心in Chinese (opens in a new tab) help.aliyun.com · checked Oct 6, 2026
- Text generation models - Qwen, DeepSeek, GLM | Model Studio - Alibaba Cloud Model Studio - Alibaba Cloud Documentation Center (opens in a new tab) alibabacloud.com · checked Oct 10, 2026
- 模型上下架与更新-大模型服务平台百炼-阿里云_大模型服务平台百炼(Model Studio)-大模型服务平台百炼(Model Studio)-阿里云帮助中心in Chinese (opens in a new tab) help.aliyun.com · checked Oct 30, 2026
- models.dev: alibaba/qwen3-coder-next (community data, MIT) (opens in a new tab) github.com · checked Sep 30, 2026
- Alibaba Cloud Model Studio pricing (opens in a new tab) alibabacloud.com · checked Oct 10, 2026
- Qwen/Qwen3-Coder-Next model card (opens in a new tab) huggingface.co · checked Oct 31, 2026
- Qwen/Qwen3-Coder-Next config.json (opens in a new tab) huggingface.co · checked Oct 31, 2026
- Qwen/Qwen3-Coder-Next files (opens in a new tab) huggingface.co · checked Nov 3, 2026
- Qwen/Qwen3-Coder-Next-FP8 files (opens in a new tab) huggingface.co · checked Nov 1, 2026
- Qwen/Qwen3-Coder-Next-GGUF files (opens in a new tab) huggingface.co · checked Nov 1, 2026
- Qwen/Qwen3-Coder-Next README.md (opens in a new tab) huggingface.co · checked Nov 4, 2026
- Qwen3-Coder-Next (opens in a new tab) huggingface.co · checked Oct 31, 2026
Alibaba Qwen’s official policies: data use, privacy, terms →
Some values on this page are still waiting to be checked against official pages: a filled dot is verified, a hollow one is not yet (with or without a source on file). How we verify data
Alibaba Qwen · China · qwen.ai
Related models
Snapshots, the same family and other models from this provider