Qwen3-30B-A3B
- In Input modalities: Text Out Output modalities: Text
- Reasoning
- Reasoning can be off
- Tool use
- Structured output
- Open weights
Prices
4 offerings · prices as listedqwen3-30b-a3b| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Standard no longer listed since Oct 11, 2026 | $0.20 | $0.80 |
| Reasoning off no longer listed since Oct 11, 2026 | — | $0.80 |
| Reasoning on no longer listed since Oct 11, 2026 | — | $2.40 |
Prices from qwen3-30b-a3b model (opens in a new tab) (checked Oct 7, 2026) · Alibaba Cloud Model Studio pricing (opens in a new tab) (checked Oct 11, 2026)
qwen3-30b-a3b| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Standard no longer listed since Oct 11, 2026 | $0.108 | $0.431 |
| Reasoning off no longer listed since Oct 11, 2026 | — | $0.431 |
| Reasoning on no longer listed since Oct 11, 2026 | — | $1.076 |
Prices from Alibaba Cloud Model Studio pricing (opens in a new tab) (checked Oct 11, 2026) · qwen3-30b-a3b model (opens in a new tab) (checked Oct 5, 2026)
qwen3-30b-a3b| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Standard no longer listed since Oct 11, 2026 | ¥0.75 | — |
| Reasoning off no longer listed since Oct 11, 2026 | — | ¥3.00 |
| Reasoning on no longer listed since Oct 11, 2026 | — | ¥7.50 |
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
- Licence
- apache-2.0
- Family
- qwen
- Model ID
Not stated on Alibaba Qwen’s pages: max input, reasoning efforts, max thinking budget, built-in tools, knowledge cutoff, aliases.
Run locally
Weights: Qwen/Qwen3-30B-A3B- Parameters 30.5B
- Active 3.3B · 128 experts, 8 per token
- Native precision BF16
- Layers 48
- KV cache per token 96 KB
- Native context 32K (model card)
- With YaRN 128K (model card)
- Config positions 40K (config.json max_position_embeddings)
- API context 128K (maker’s API)
- Needs
- ≈ 18.9 GB (weights 17.3 (maker’s file) + KV cache 0.8 + overhead 0.9)
- Usable
- 21.6 GB (90% of its memory)
- Generation speed
- 179–269 tok/s (reads the active 3.3B per token)
- Longest context
- 37K (longer doesn’t fit)
- First token
- about 0.11–0.16 s for a 2K prompt
| Precision | Weights | Needs at 8K | On your hardware | Longest context |
|---|---|---|---|---|
| BF16 (native)maker’s file | 56.9 GB | ≈ 60.5 | ○Needs system memory 12–17 tok/s | — |
| Q8_0maker’s file | 30.3 GB | ≈ 32.5 | ○Needs system memory 30–44 tok/s | — |
| FP8maker’s file | 30.2 GB | ≈ 32.5 | ○Needs system memory 30–45 tok/s | — |
| Q6_Kmaker’s file | 23.4 GB | ≈ 25.3 | ○Needs system memory 69–103 tok/s | — |
| Q5_K_Mmaker’s file | 20.2 GB | ≈ 22.0 | ○Needs system memory 177–265 tok/s | — |
| Q5_0maker’s file | 19.6 GB | ≈ 21.4 | ●Runs well 163–245 tok/s | 10K |
| Q4_K_Mmaker’s file | 17.3 GB | ≈ 18.9 | ●Runs well 179–269 tok/s | 37K |
| Q4_0 | 16.0 GB | ≈ 17.5 | ●Runs well 189–284 tok/s | 128K |
| INT4 (AWQ / GPTQ)maker’s file | 15.8 GB | ≈ 17.3 | ●Runs well 191–287 tok/s | 128K |
| IQ4_XS | 15.1 GB | ≈ 16.6 | ●Runs well 197–295 tok/s | 128K |
| Q3_K_Mquality drops | 13.9 GB | ≈ 15.3 | ●Runs well 208–313 tok/s | 128K |
| IQ3_XXSquality drops | 10.9 GB | ≈ 12.2 | ●Runs well 244–366 tok/s | 128K |
| Q2_Kquality drops | 10.7 GB | ≈ 11.9 | ●Runs well 247–370 tok/s | 128K |
| IQ2_XXSquality drops | 7.3 GB | ≈ 8.6 | ●Runs well 305–457 tok/s | 128K |
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 GeForce RTX 4090; unified memory Apple M6 (24GB or 32GB) · 32 GB.
Run commands
llama-server -hf Qwen/Qwen3-30B-A3B-GGUF:Q4_K_M -c 8192
ollama run hf.co/Qwen/Qwen3-30B-A3B-GGUF:Q4_K_M
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 us- Epoch Capabilities Index 134.3
#109 of 118 listed models · CI 131.4–136.9
- LMArena Text 1,327
#136 of 163 listed models · CI 1,322–1,331 · 26,156 votes
- Epoch AI · GPQA Diamond 61.7%
#109 of 168 listed models · ±2.97
Also listed: reasoning off 50.4%
- Epoch AI · OTIS Mock AIME 2024–2025 62.8%
#100 of 159 listed models · ±6.21
Also listed: reasoning off 25.6%
- Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Nov 4 – 5, 2026
- LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Oct 28, 2026
- EQ-Bench by Samuel J. Paech (eqbench.com; EQ-bench/EQ-bench-site on GitHub), MIT License. · License · as of Oct 28, 2026
- changes we made
Changes
All changes →- Now priced on Cloudflare Workers AI Input $0.051 · Output $0.335
- Context window 256K → 131.07K
- Max output 8.19K → 16.38K
Sources
- Alibaba Cloud Model Studio pricing (opens in a new tab) alibabacloud.com · checked Oct 11, 2026
- 阿里云百炼：模型价格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
- qwen3-30b-a3b model (opens in a new tab) alibabacloud.com · checked Oct 5, 2026
- qwen3-30b-a3b model (opens in a new tab) alibabacloud.com · checked Oct 7, 2026
- Cloudflare Workers AI pricing (opens in a new tab) developers.cloudflare.com · checked Oct 31, 2026
- Qwen/Qwen3-30B-A3B model card (opens in a new tab) huggingface.co · checked Nov 1, 2026
- Qwen/Qwen3-30B-A3B config.json (opens in a new tab) huggingface.co · checked Nov 1, 2026
- Qwen/Qwen3-30B-A3B files (opens in a new tab) huggingface.co · checked Nov 2, 2026
- Qwen/Qwen3-30B-A3B-FP8 files (opens in a new tab) huggingface.co · checked Oct 31, 2026
- Qwen/Qwen3-30B-A3B-GGUF files (opens in a new tab) huggingface.co · checked Oct 31, 2026
- Qwen/Qwen3-30B-A3B-GPTQ-Int4 files (opens in a new tab) huggingface.co · checked Nov 2, 2026
- Qwen/Qwen3-30B-A3B-MLX-4bit files (opens in a new tab) huggingface.co · checked Nov 2, 2026
- Qwen/Qwen3-30B-A3B-MLX-6bit files (opens in a new tab) huggingface.co · checked Oct 31, 2026
- Qwen/Qwen3-30B-A3B-MLX-8bit files (opens in a new tab) huggingface.co · checked Nov 3, 2026
- Qwen/Qwen3-30B-A3B-MLX-bf16 files (opens in a new tab) huggingface.co · checked Oct 31, 2026
- Qwen/Qwen3-30B-A3B README.md (opens in a new tab) huggingface.co · checked Nov 4, 2026
- Qwen/Qwen3-30B-A3B model card (opens in a new tab) huggingface.co · checked Oct 31, 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
Alibaba Qwen’s official policies: data use, privacy, terms →
Alibaba Qwen · China · qwen.ai
Related models
Snapshots, the same family and other models from this provider