Qwen3.5 35B A3B Base — Hardware Requirements & GPU Compatibility
Vision
Qwen3.5-35B-A3B-Base is the mixture-of-experts member of Alibaba's Qwen3.5 base family, pairing a 256-expert MoE layer (8 routed plus 1 shared expert per token) with the same Gated DeltaNet and gated-attention hybrid backbone used across the line. It totals roughly 36 billion parameters but activates only about 3 billion per token (the "A3B" in its name), so decoding stays fast even though the full expert set must stay resident in memory. Like its dense siblings it is pretrain-only, for fine-tuning and research, not direct chat; unlike them, its card lists only a pretraining stage, with no post-training pass. It needs a high-end consumer GPU once quantized. Context length is 262,144 tokens natively, extensible to 1,010,000 tokens. It is released under the Apache 2.0 license, permitting unrestricted commercial and research use, and was published in February 2026, alongside four smaller dense Qwen3.5-Base models from 0.8B to 9B parameters.
135.9K downloads
147 likesSpecifications
- Publisher
- Alibaba
- Family
- Qwen 3.5
- Parameters
- 36.0B
- Architecture
- Qwen3_5MoeForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 248,320
- Release Date
- 2026-02-24
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does Qwen3.5 35B A3B Base Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 15.7 GB | 26.3 GB | 15.28 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 16.1 GB | 26.8 GB | 15.73 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 17.9 GB | 28.6 GB | 17.53 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 18.4 GB | 29.0 GB | 17.98 GB | 4-bit legacy quantization |
| Q4_K_M | 4.80 | 21.9 GB | 32.6 GB | 21.57 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 26 GB | 36.6 GB | 25.62 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 30.0 GB | 40.7 GB | 29.66 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 36.3 GB | 47.0 GB | 35.95 GB | 8-bit quantization, near-lossless |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run Qwen3.5 35B A3B Base?
Q4_K_M · 21.9 GBQwen3.5 35B A3B Base (Q4_K_M) requires 21.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 29+ GB is recommended. Using the full 262K context window can add up to 10.7 GB, bringing total usage to 32.6 GB. 7 GPUs can run it, including NVIDIA GeForce RTX 5090.
All compatible consumer-level GPUs are running near their VRAM limit. You may also want to consider professional GPUs (e.g., NVIDIA A100, H100) which offer significantly more VRAM. For more headroom and better throughput, consider a multi-GPU configuration with tensor parallelism (supported by tools like vLLM, llama.cpp, or text-generation-inference).
Which Devices Can Run Qwen3.5 35B A3B Base?
Q4_K_M · 21.9 GB41 devices with unified memory can run Qwen3.5 35B A3B Base, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Mini M4 Pro (24 GB).
Runs great
— Plenty of headroomDecent
— Enough memory, may be tightWhere to Download Qwen3.5 35B A3B Base
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Related Models
Frequently Asked Questions
- How much VRAM does Qwen3.5 35B A3B Base need?
Qwen3.5 35B A3B Base requires 21.9 GB of VRAM at Q4_K_M, or 72.3 GB at BF16. Full 262K context adds up to 10.7 GB (32.6 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 36.0B × 4.8 bits ÷ 8 = 21.6 GB
KV Cache + Overhead ≈ 0.3 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 11 GB (at full 262K context)
VRAM usage by quantization
Q4_K_M21.9 GBQ4_K_M + full context32.6 GB- Can NVIDIA GeForce RTX 4090 run Qwen3.5 35B A3B Base?
Yes, at Q4_K_M (21.9 GB) or lower. Higher quantizations like Q5_K_S (25.1 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
- What's the best quantization for Qwen3.5 35B A3B Base?
For Qwen3.5 35B A3B Base, Q4_K_M (21.9 GB) offers the best balance of quality and VRAM usage. Q5_K_S (25.1 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 10.3 GB.
VRAM requirement by quantization
IQ2_XXS10.3 GBIQ3_XS15.2 GBQ3_K_M17.9 GBQ4_K_M ★21.9 GBQ5_K_S25.1 GBBF1672.3 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run Qwen3.5 35B A3B Base on a Mac?
Qwen3.5 35B A3B Base requires at least 10.3 GB at IQ2_XXS, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Qwen3.5 35B A3B Base locally?
Yes — Qwen3.5 35B A3B Base can run locally on consumer hardware. At Q4_K_M quantization it needs 21.9 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Qwen3.5 35B A3B Base?
At Q4_K_M, Qwen3.5 35B A3B Base can reach ~116 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~151 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 21.9 × 0.65 = ~363 tok/s
Estimated speed at Q4_K_M (21.9 GB)
~363 tok/s~151 tok/s~363 tok/s~322 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Qwen3.5 35B A3B Base?
At Q4_K_M, the download is about 21.57 GB. The full-precision BF16 version is 71.90 GB. The smallest option (IQ2_XXS) is 9.89 GB.
- Which GPUs can run Qwen3.5 35B A3B Base?
7 consumer GPUs can run Qwen3.5 35B A3B Base at Q4_K_M (21.9 GB). Top options include AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090.
- Which devices can run Qwen3.5 35B A3B Base?
41 devices with unified memory can run Qwen3.5 35B A3B Base at Q4_K_M (21.9 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.