Leaderboard / Alibaba (Qwen)

Qwen3-Coder-Next

Name: Qwen3-Coder-Next
Price: 0.29 USD
Author: Alibaba (Qwen)

Alibaba (Qwen) · Qwen3 · released 2026-01-30 · data checked 2026-11-01 Needs verification

Qwen3-Coder-Next is a 79.7B-parameter Mixture-of-Experts model (3B active per token) from Alibaba (Qwen) (China), released 2026-01-30 under the Apache-2.0 license. It doesn't have enough reported benchmarks for an index score yet. It needs about 50.6 GB of memory at 4-bit with an 8K-token context and costs $0.29 per million tokens through the cheapest API provider.

Coding-agent model (80B total, 3B active). SWE-bench ranges 70-74 depending on the agent scaffold.

Small LM Index

—

Needs 2 core benchmarks (has 1)

Parameters

79.7B

3B active · MoE

Price per 1M tokens

$0.29

$0.12 in / $0.80 out · blended 3:1

Context window

262K

tokens

Benchmarks

Each bar shows this model's score. The tick marks the best large model on that benchmark. Sources: 2 reported by Alibaba (Qwen). How we label sources.

SWE-bench Verified Lab-reported Coding70.6

Terminal-Bench Lab-reported Agentic · Terminal-Bench 2.036.2

Class best: Nemotron 3.5 Bolt 74B-A9B 53.9

Specifications

Lab: Alibaba (Qwen) (CN · China)
Architecture: Mixture-of-Experts
Total parameters: 79.7B
Active parameters: 3B
Size class: Large
Context window: 262K tokens
Input modalities: text
Reasoning: Instruct (no thinking)
License: Apache-2.0
Openness: Open weights
Memory to run: ~50.6 GB at 4-bit, 8K context (more sizes)
Hugging Face: 431.4K downloads (30 days) · 1.7K likes
API providers: 1 on OpenRouter
API price: $0.12 in / $0.80 out per 1M (OpenRouter, 2026-11-03)

Weights on Hugging Face ↗Try on OpenRouter ↗

Compare with

Qwen3-Coder-Next vs Qwen3-Next 80B-A3B Thinking (80B, index 63.3)
Qwen3-Coder-Next vs Kolibri-1 (78.1B, index 68.3)
Qwen3-Coder-Next vs Nemotron 3.5 Bolt 74B-A9B (74B, index 73.3)

Sources

Something wrong here? Report it.

Run it locally

Uses unsloth/Qwen3-Coder-Next-GGUF (Q4_K_M). Swap the tag for another size: BF16, IQ4_NL, IQ4_XS, MXFP4, Q2_K, Q2_K_L, Q3_K_M, Q3_K_S….

Ollama

ollama run hf.co/unsloth/Qwen3-Coder-Next-GGUF:Q4_K_M

llama.cpp

llama-server -hf unsloth/Qwen3-Coder-Next-GGUF:Q4_K_M

Then open http://localhost:8080 for a chat window.

MLX (Mac)

pip install -U mlx-lm
mlx_lm.chat --model mlx-community/Qwen3-Coder-Next-4bit

vLLM (GPU server)

vllm serve Qwen/Qwen3-Coder-Next

Full-precision weights, for a server GPU.

Open in LM Studio

Memory needed: weights, context cache and runtime buffers.

Quantization	4K context	32K context	256K context
16-bit F16	162 GB	162 GB	168 GB
8-bit Q8_0	87 GB	88 GB	93 GB
4-bit Q4_K_M	51 GB	51 GB	57 GB

Check it against your GPU, Mac or phone →