Alibaba Cloud, released 2025-04-28. Sparse MoE decoder, 30B total, 3B active, 128K context. Attention: GQA.
Sparse MoEAlibaba Cloud 🇨🇳 · July 31, 2025
Qwen3 Coder Flash (30B-A3B)
Coding-tuned Qwen model that keeps a straightforward grouped-query MoE stack instead of the newer hybrid-attention variants.
Key detail
Uses 128 experts with 8 active per token and a native 256k context window for coding workloads.Total params
30B
Active / token
3.3B
Context
256K
Fact sheet
- Scale
- 30B total, 3.3B active (11% active)
- Context window
- 256,000 tokens
- Decoder type
- Sparse MoE
- Attention
- GQA
- Layer mix
- 48 GQA
- KV cache / token (bf16)
- 96 KiB · Moderate
- License
- Apache License 2.0
- Released
Efficiency
Parameters active per token
11%
3.3B of 30B weights run for each token.
AA Intelligence Index v4.3.2
Source9.6
- Scientific
- 1.9
Architecture concepts
Sources & resources
Use the ← and → keys to step through releases.
Similar architectures
Sarvam AI, released 2026-03-03. Sparse MoE decoder, 30B total, 2.4B active, 128K context. Attention: GQA with QK-Norm.
Cohere, released 2026-06-05. Sparse MoE decoder, 30B total, 3B active, 256K context. Attention: 8:1 GQA with 3:1 sliding-window/global attention.
Poolside, released 2026-04-28. Sparse MoE decoder, 33B total, 3B active, 128K context. Attention: Gated GQA with QK-Norm and 3:1 sliding-window/global attention.