BestLLMfor EN Your hardware. Your LLM. Your call.
APIOpen data Find my LLM
Model fiche

Qwen3-Coder-Next 80B-A3B

By Alibaba · China

Updated 2026-07-13

code moe
Parameters
80B
License
Apache 2.0
Context
255k
VRAM (Q4)
48 GB
Released
February 2026

Overview

Alibaba's hybrid Gated DeltaNet + Attention MoE with 80B total and 3B active parameters. Purpose-built as a local coding copilot that fits on a 24GB GPU.

When to pick this model

  • Local Copilot-style code completion
  • Long-context refactoring up to 262K tokens
  • IDE plugins running on consumer hardware
  • Apache-licensed commercial code tooling
  • Reducing reliance on cloud coding APIs

VRAM requirements by quantization

VRAM REQUIRED (GB)121624324880128Q4_K_M48 GBQ5_K_M58 GBQ8_086 GBFP16160 GB
QuantizationVRAM required
Q4_K_M (recommended)48 GB
Q5_K_M58 GB
Q8_086 GB
FP16 (no quantization)160 GB

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

In practice, Qwen3-Coder-Next 80B-A3B spills past single consumer GPUs even at Q4_K_M (48 GB) — think dual-GPU or workstation cards. Stepping up to Q8_0 nearly doubles the footprint to 86 GB, and unquantized FP16 weights take 160 GB — plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity.

Without a GPU, Qwen3-Coder-Next 80B-A3B needs roughly 72 GB of system RAM to run on CPU via llama.cpp or Ollama — workable for background jobs, but far slower than GPU inference. Throughput estimates from our compatibility engine: around 5 tokens/sec on entry-level GPUs, on the order of 18 tokens/sec on a mid-range card, and up to 50 tokens/sec on high-end hardware — assuming the chosen quantization fully fits in VRAM.

What hardware do you need

The table below matches Qwen3-Coder-Next 80B-A3B to common GPU memory tiers, using the highest-fidelity quantization that fully fits each card class. Spilling layers to system RAM works but costs most of the speed, so size your card to the quantization you actually want to run.

GPU memoryExample cardsBest fit for Qwen3-Coder-Next 80B-A3B
8 GBRTX 5070 Laptop, RTX 5060, RTX 5060 Ti 8GBDoes not fit — needs 48 GB at Q4_K_M
12 GBRTX 5070, RTX 5070 Ti Laptop, RTX 4080 LaptopDoes not fit — needs 48 GB at Q4_K_M
16 GBRTX 5080, RTX 4080 Super, Radeon RX 9070 XTDoes not fit — needs 48 GB at Q4_K_M
24 GBRTX 4090, Radeon RX 7900 XTX, RTX 5090 LaptopDoes not fit — needs 48 GB at Q4_K_M
32 GBRTX 5090Does not fit — needs 48 GB at Q4_K_M

Which GPU should you buy to run Qwen3-Coder-Next 80B-A3B?

To run Qwen3-Coder-Next 80B-A3B locally at Q4, you need ~48 GB of VRAM. The best value for this is a Apple Mac Studio (64+ GB unified memory).

Check Apple Mac Studio price on Amazon →

As an Amazon Associate, BestLLMfor earns from qualifying purchases, at no extra cost to you. It does not influence our independent rankings.

Strengths

  • Runs as a local copilot on a 24GB GPU
  • 262K context fits entire codebases
  • Hybrid architecture keeps memory low
  • Apache 2.0 license

Limitations

  • Hybrid architecture means partial llama.cpp support
  • Less mature than dense coder alternatives
  • Tooling lags behind standard transformer models

Typical workloads

In our catalog grid, Qwen3-Coder-Next 80B-A3B is filed under Local Copilot, Senior Code, Refactor — the use cases where its size/quality trade-off makes the most sense. Its tags translate to concrete workloads: code generation and review (pair it with an editor integration like Continue.dev or Cline).

The 255k-token context window is large enough to hold entire codebases' worth of files or long reports in a single prompt, which is what makes local RAG and document analysis practical. The Apache 2.0 license is permissive, so shipping it inside a commercial product raises no special legal questions.

Architecture & training

Architecture: MoE 80B/3B · Gated DeltaNet + hybrid Attention · 262k ctx

Training: Agentic code specialist.

Verdict

Choose this when you want a local Copilot replacement and can tolerate early-stage tooling friction.

Quick start

ollama run qwen3-coder-next

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.

Similar models worth comparing

Frequently asked questions

How much VRAM does Qwen3-Coder-Next 80B-A3B need?

At the recommended Q4_K_M quantization, Qwen3-Coder-Next 80B-A3B needs about 48 GB of VRAM. Q8_0 takes 86 GB, and unquantized FP16 weights take 160 GB.

Can Qwen3-Coder-Next 80B-A3B run without a GPU?

Yes — with roughly 72 GB of system RAM it runs CPU-only through llama.cpp or Ollama. Expect a fraction of GPU speed, which is fine for background or batch jobs but slow for interactive chat.

What context window does Qwen3-Coder-Next 80B-A3B support?

Qwen3-Coder-Next 80B-A3B supports a 255k-token context window (262,000 tokens).

Can I use Qwen3-Coder-Next 80B-A3B commercially?

Yes. Qwen3-Coder-Next 80B-A3B is released under Apache 2.0, a permissive open-source license that allows commercial use, modification and redistribution.

How fast is Qwen3-Coder-Next 80B-A3B on consumer hardware?

Our compatibility engine estimates on the order of 18 tokens/sec on a mid-range GPU and up to 50 tokens/sec on high-end cards, assuming the quantization fully fits in VRAM.

Which quantization of Qwen3-Coder-Next 80B-A3B should I download first?

Start with Q4_K_M (48 GB) — the standard size/quality sweet spot. Step up to Q5_K_M or Q8_0 only if you have VRAM headroom. It does not fit a single 24 GB consumer card — plan for multi-GPU or server hardware.

Tools

Is Qwen3-Coder-Next 80B-A3B the right pick for you?

Compute self-hosted ROI → Back to catalog