[whatmodel.sh](../../README.md) / Qwen3-Coder 480B-A35B Instruct # Qwen3-Coder 480B-A35B Instruct Alibaba · released 2025-07 · 480B params (35B active) · apache-2.0 **score 63.3** ±1.2 · #166 of 732 · from 10 benchmark results across 4 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at _{● marks Qwen3-Coder 480B-A35B Instruct among every model on that task or test, worst to best.}

task            vs the others          rank  counts  tests
agent tasks     ▁▁▃▆█▄●▅▄▄▂▁    #143 of 313     33%      7
coding          ▁▁▁▃▅▇█●▅▄▃▁    #152 of 469     65%      6
chat & writing  ▁▁▁▃▆▄▆█▆●▇▃   #129 of 402*     14%      1

agent tasks · #143 of 313 · 33% of its score

Beats [o3](o3.md), [Gemini 2.5 Pro](gemini-2-5-pro.md), [DeepSeek V3.2 Exp](deepseek-v3-2-exp.md) here, though they rank higher overall. Trails [SmolLM3 3B](smollm3-3b.md), [Jamba 1.5 Large](jamba-1-5-large.md), [GPT-4o (2024-05-13)](gpt-4o-2024-05-13.md) here, though they rank lower overall.

test                                result  vs the others in its run    difficulty            counts  why
Terminal-Bench 2.0 with Terminus …   23.9%  ▃··▃·▃▃●█▅▃▅ beats 4/14     very hard · avg 6.1%    9.4%  sharp
SWE-bench Verified (maker)           67.0%  ▁▁▁▁▁▁▂▄●▆▁▁ beats 78/138   hard · avg 28%          7.8%  sharp · at its level · small run
τ-bench Retail (maker)               77.5%  ▁····▃·▂▇█▂● beats 23/28    medium · avg 46%        5.3%  small run
SWE-bench Multilingual (maker)       54.7%  ▁··▁▃▁●██▂▂▁ beats 16/65    hard · avg 24%          5.0%  at its level · small run
τ-bench Airline (maker)              60.0%  ▁▁▁·▅▄▁█▇▁▁● beats 26/28    hard · avg 29%          4.2%  at its level · small run
BFCL v3 (maker)                      68.7%  ▁▁▁▁▁▃▅█▆●▄▁ beats 144/171  medium · avg 56%        1.6%  blunt · at its level · small run
SWE-bench Live (maker)               26.3%  █··········● beats 1/1                                 —  not counted yet: no other model in the …

coding · #152 of 469 · 65% of its score

Beats [Qwen3-Coder-Next (80B-A3B)](qwen3-coder-next.md), [Gemini 2.5 Pro](gemini-2-5-pro.md), [Gemini 3.1 Flash-Lite](gemini-3-1-flash-lite.md) here, though they rank higher overall. Trails [Ministral 3 8B Reasoning 2512](ministral-3-8b-reasoning.md), [Phi-4-reasoning 14B](phi-4-reasoning.md), [Llama 3.1 Nemotron Ultra 253B v1](llama-3-1-nemotron-ultra-253b.md) here, though they rank lower overall.

test                                result  vs the others in its run   difficulty           counts  why
Arena WebDev (Code Arena)           1274.8  ▁▁▃●▅█▃▆▅▅▁▂ beats 18/129  avg 981                 25%  sharp
Aider polyglot (225 exercises, tw…   61.8%  ▂▁▁▃▆▃▄▆●▃▄▂ beats 55/71   very hard · avg 10%     19%  sharp · at its level · small run
WebApp1K-Duo-React pass@1 (onekq-…   61.2%  ▁··▁·▁▁▁▆●▇█ beats 15/45   hard · avg 29%         8.0%  at its level
SWE-bench Verified (maker)           67.0%  ▁▁▁▁▁▁▂▄●▆▁▁ beats 78/138  hard · avg 28%         7.8%  sharp · at its level · small run
SWE-bench Multilingual (maker)       54.7%  ▁··▁▃▁●██▂▂▁ beats 16/65   hard · avg 24%         5.0%  at its level · small run
SWE-bench Live (maker)               26.3%  █··········● beats 1/1                               —  not counted yet: no other model in the …

chat & writing · #129 of 402 · 14% of its score

Beats [LongCat 2.0 (1.6T-A48B)](longcat-2-0.md), [o3-mini](o3-mini.md), [MiMo-V2-Flash (309B-A15B)](mimo-v2-flash.md) here, though they rank higher overall. Trails [LFM2.5 2.6B](lfm2-5-2-6b.md), [Nemotron-Cascade-8B](nemotron-cascade-8b.md), [DeepSeek V3 0324](deepseek-v3-0324.md) here, though they rank lower overall.

test            result  vs the others in its run    difficulty  counts  why
LMArena (text)  1387.6  ▁▁▁▂▃▃▃▆▆●█▃ beats 229/385  avg 1260       14%  blunt

## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its *counts* column above; here it is added up by who ran the tests. [Method](../method.md)

source                                    results  share of the score
Qwen/Qwen3-Coder-30B-A3B-Instruct model…        6   43% ▇▇▇▇▇
LMArena leaderboard dataset                     2   39% ▇▇▇▇▇
Terminal-Bench: Benchmarking Agents on …        1    9% ▇
WebApp1K models leaderboard                     1    8% ▇
prior: pull toward the middle                        1% ▇
weighted average                                   = 63.3

_{Licences: Qwen/Qwen3-Coder-30B-A3B-Instruct model card ([Apache-2.0](../../licenses/card-qwen-qwen3-coder-30b-a3b-instruct.txt)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Terminal-Bench: Benchmarking Agents on Hard ([CC BY 4.0](../../licenses/paper-2601.11868v1-CC-BY-4.0.txt)) · WebApp1K models leaderboard ([MIT](../../licenses/webapp1k-leaderboard-MIT.txt))}

Not counted (8)

source                  test                            score  why
Qwen/Qwen3-Coder-480B…  Aider polyglot (225 exercises…  61.8%  only model in this source's run
Qwen/Qwen3-Coder-480B…  BFCL v3                         68.7%  only model in this source's run
Qwen/Qwen3-Coder-480B…  SWE-bench Live                  26.3%  too few models have it yet
Qwen/Qwen3-Coder-480B…  SWE-bench Multilingual          54.7%  only model in this source's run
Qwen/Qwen3-Coder-480B…  SWE-bench Verified              69.6%  only model in this source's run
Qwen/Qwen3-Coder-480B…  τ-bench Airline                 60.0%  only model in this source's run
Qwen/Qwen3-Coder-480B…  τ-bench Retail                  77.5%  only model in this source's run
Qwen/Qwen3-Coder-30B-…  SWE-bench Live                  26.3%  too few models have it yet

## How to run it ### Through an API api (Novita AI) · $0.67/M blended · $0.38 in / $1.55 out per 1M tokens · [novita.ai](https://novita.ai/models) _{[price source](https://github.com/BerriAI/litellm/blob/c19ce71bcdad234ca87a7af10cee45375ce811a8/model_prices_and_context_window.json): USD per 1M text tokens, standard uncached input and output. Model endpoint qwen/qwen3-coder-480b-a35b-instruct. Product: Serverless chat completions. API model: qwen/qwen3-coder-480b-a35b-instruct. Settings: service tier=standard, batch=no, cache=uncached input. Additional/excluded charges: Tools, cache storage, priority service and other modalities excluded.. Currency: USD. · [context source](https://github.com/BerriAI/litellm/blob/c19ce71bcdad234ca87a7af10cee45375ce811a8/model_prices_and_context_window.json): provider maximum input tokens; not an intrinsic model maximum or legacy max\_tokens} ### llama.cpp (recommended) 1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md) ```sh # macOS / Linux curl -LsSf https://llama.app/install.sh | sh # Windows irm https://llama.app/install.ps1 | iex ``` 2. Run Qwen3-Coder 480B-A35B Instruct (ud-q2_k_xl, needs ~144.5 GB, from [huggingface.co/unsloth](https://huggingface.co/unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF/tree/main)) ```sh llama cli -hf unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF:UD-Q2_K_XL ``` 3. Or start a local server and open the address it prints (http://localhost:9931) ```sh llama serve -hf unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF:UD-Q2_K_XL ``` Other versions: ud-iq1_m (~72.5 GB) `llama cli -hf unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF:UD-IQ1_M` ### LM Studio 1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download) 2. Run Qwen3-Coder 480B-A35B Instruct (4bit, needs ~270.1 GB, Mac only, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/Qwen3-Coder-480B-A35B-Instruct-MLX-4bit/tree/a13176858e9594b24a5ab04298d6d0939e51ab66)) ```sh lms get https://huggingface.co/lmstudio-community/Qwen3-Coder-480B-A35B-Instruct-MLX-4bit -y ``` 3. Then run lms ls to see its name, and lms chat followed by that name Other versions: ud-q2_k_xl (~144.5 GB) `lms get https://huggingface.co/unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF@UD-Q2_K_XL -y` · ud-iq1_m (~72.5 GB) `lms get https://huggingface.co/unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF@UD-IQ1_M -y` · 8bit (~510.2 GB, Mac only) `lms get https://huggingface.co/lmstudio-community/Qwen3-Coder-480B-A35B-Instruct-MLX-8bit -y` ### Ollama 1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download) ```sh # Linux curl -fsSL https://ollama.com/install.sh | sh ``` 2. Run Qwen3-Coder 480B-A35B Instruct (Q4_K_M, needs ~288.5 GB, from [ollama.com](https://ollama.com/library/qwen3-coder/tags)) ```sh ollama run qwen3-coder:480b ``` ### MLX (Mac) 1. Install MLX (Mac). Apple silicon Macs only, text only. ```sh # with uv uv tool install mlx-lm # or pip install mlx-lm ``` 2. Run Qwen3-Coder 480B-A35B Instruct (4bit, needs ~270.1 GB, Mac only, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/Qwen3-Coder-480B-A35B-Instruct-MLX-4bit/tree/a13176858e9594b24a5ab04298d6d0939e51ab66)) ```sh mlx_lm.chat --model lmstudio-community/Qwen3-Coder-480B-A35B-Instruct-MLX-4bit ``` 3. Or start an OpenAI-compatible server at http://localhost:8080/v1 ```sh mlx_lm.server --model lmstudio-community/Qwen3-Coder-480B-A35B-Instruct-MLX-4bit ``` Other versions: 8bit (~510.2 GB, Mac only) `mlx_lm.chat --model lmstudio-community/Qwen3-Coder-480B-A35B-Instruct-MLX-8bit` Model repository and documentation: [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct) [interactive version](https://whatmodel.sh/model/qwen3-coder-480b-a35b) · [JSON](https://whatmodel.sh/api?format=json&all=1)