Can I run Qwen3-Coder-Next locally?

Yes, the weights are open and available on Hugging Face at Qwen/Qwen3-Coder-Next. Ollama, LM Studio, llama.cpp, MLX-LM and KTransformers all support it, though the 80B total parameter count means you will need substantial hardware.

What does it cost via API, and how long is the context?

Pricing is $0.12 per million input tokens and $0.80 per million output tokens, with a native context window of 262,144 tokens.

How is this different from other Qwen3 models?

Qwen3-Coder-Next is purpose-built for coding agents rather than general use: it runs in non-thinking mode only, skips chain-of-thought blocks entirely, and its training emphasises tool use, execution-failure recovery and long-horizon agentic tasks over broad conversational ability.

Qwen3 Coder Next

5847°

#llm #qwen #coding #agents #long context #local

Run this model

Offers / Ads

OpenRouter Terminal / API

Pay per token

Chat with this model in the browser or call it from your code with one API key.

Qwen | 262,144 token context | $0.12/M input, $0.80/M output. Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows. It uses a sparse MoE design with 80B total parameters and only 3B activated per token, delivering performance comparable to models with 10 to 20x higher active compute, which makes it well suited for cost-sensitive, always-on agent deployment. The model is trained with a strong agentic focus and performs reliably on long-horizon coding tasks, complex tool usage, and recovery from execution failures. With a native…Qwen | 262,144 token context | $0.12/M input, $0.80/M output. Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows. It uses a sparse MoE design with 80B total parameters and only 3B activated per token, delivering performance comparable to models with 10 to 20x higher active compute, which makes it well suited for cost-sensitive, always-on agent deployment. The model is trained with a strong agentic focus and performs reliably on long-horizon coding tasks, complex tool usage, and recovery from execution failures. With a native 256k context window, it integrates cleanly into real-world CLI and IDE environments and adapts well to common agent scaffolds used by modern coding tools. The model operates exclusively in non-thinking mode and does not emit blocks, simplifying integration for production coding agents.

Guides for this model

Ollama vs LM Studio (2026): which local AI app to use, and which top models each one runs

How much VRAM do I actually need for local AI?

What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)

All guides

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

Qwen3-Coder-Next is an open-weight MoE model from Qwen built specifically for coding agents, with 80B total parameters but only 3B active per token. That efficiency makes it viable for always-on agent deployment at a fraction of the compute cost of comparably capable models. If you run coding agents in a CLI or IDE and want something you can host yourself, this is worth a serious look.

Known for

MoE design: 80B total, 3B active per token — low inference cost for its capability tier.
Native 256k context handles large codebases and long multi-step agent sessions.
Trained for agentic work: tool use, long-horizon tasks, and recovering from execution failures.
Integrates with Claude Code, Cline, LM Studio, Ollama, llama.cpp and similar scaffolds.
No thinking-mode overhead — clean, direct output suited to production agent pipelines.

Watch out for

Non-thinking mode only; if you need chain-of-thought reasoning blocks, look elsewhere.
80B total parameters means local deployment requires serious hardware despite the low active count.
Maker has not published a detailed licence summary on the model page — check the repo before commercial use.

Best for

Prompts to try

Debug a failing agent tool

I have a Python coding agent that calls a shell tool to run pytest. The tool returns exit code 1 but the agent keeps retrying the same command without reading the error output. Write the tool-call handler and the retry logic so the agent parses stderr, identifies the failing test, and attempts a targeted fix before retrying.

Refactor a large module

Here is a 400-line Python module that mixes database access, business logic and HTTP serialisation in the same file. Split it into three files following a clean layered architecture, keep all existing function signatures public, and add a brief comment at the top of each new file explaining its responsibility.

Write a CLI scaffolding tool

Write a self-contained Python CLI tool using argparse that scaffolds a new FastAPI project. It should create a directory structure with app, models, routers and tests folders, generate a minimal main.py, a requirements.txt with fastapi and uvicorn, and a README with a one-command run instruction.

Compare with

Qwen3.8 Max (0902)
API-only model
1,000,000 token context · $2.00/M input, $6.00/M output
Qwen3.7 Plus
API-only model
1,000,000 token context · $0.32/M input, $1.28/M output
Qwen3.6 Plus
API-only model
1,000,000 token context · $0.33/M input, $1.95/M output
Qwen3.6 Max Preview
API-only model
262,144 token context · $1.03/M input, $6.16/M output

More text models from qwen

Questions people ask

Can I run Qwen3-Coder-Next locally?: Yes, the weights are open and available on Hugging Face at Qwen/Qwen3-Coder-Next. Ollama, LM Studio, llama.cpp, MLX-LM and KTransformers all support it, though the 80B total parameter count means you will need substantial hardware.
What does it cost via API, and how long is the context?: Pricing is $0.12 per million input tokens and $0.80 per million output tokens, with a native context window of 262,144 tokens.
How is this different from other Qwen3 models?: Qwen3-Coder-Next is purpose-built for coding agents rather than general use: it runs in non-thinking mode only, skips chain-of-thought blocks entirely, and its training emphasises tool use, execution-failure recovery and long-horizon agentic tasks over broad conversational ability.

Captain's verdict
A rare open-weight model that takes agent deployment seriously — the MoE efficiency makes always-on hosting genuinely practical.

Model card on Hugging Face

Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.

Ranking badge

Updated Oct 22, 2026

Made this model? Show this badge on your page.

Paste it into your Hugging Face card or Civitai description.

Change colors / personalize / get code

Not this creator?

Get your own badge

How this score adds up

This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.

Glossary: the terms used here

Heat score: One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor: The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents: Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale: Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes: Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate: How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated: Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder: Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde: A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites: Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens: A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint): The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding: Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights: The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age: Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost: A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score: A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period): A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend: How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model: The middle of the catalog: half the models with a figure grew faster over the period, half slower.

Similar Models

AI Models for Image, Video & Text Generation

Qwen3 Coder Next is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.

More Text / LLM models llm qwen coding agents long context local

Browse by Model Family