[whatmodel.sh](../../README.md) / Llama 4 Maverick (402B-A17B)
# Llama 4 Maverick (402B-A17B)
Meta · released 2025-04 · 402B params (17B active) · llama-4
**score 57.2** ±0.8 · #241 of 732 · from 27 benchmark results across 20 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Llama 4 Maverick (402B-A17B) among every model on that task or test, worst to best.
task vs the others rank counts tests
agent tasks ▁▁▃●█▄▅▅▄▄▂▁ #252 of 313 9.1% 5
coding ▁▁▁▃▅▇●▇▅▄▃▁ #221 of 469 17% 3
chat & writing ▁▁▁▃▆▄▆●▆▇▇▃ #197 of 402 14% 3
math ▁▁▃▅█●▄█▄▂▁▁ #203 of 431 19% 3
vision & documents ▁▁▂▂▁▃●▂▆▇█▄ #101 of 153 11% 4
web search ▁▁▁▅●█▃▄·▁·▁ #57 of 84* 2.7% 1
agent tasks · #252 of 313 · 9.1% of its score
Trails [Granite 4.0 1B](granite-4-0-1b.md), [SmolLM3 3B](smollm3-3b.md), [Qwen3-VL 2B Instruct](qwen3-vl-2b-instruct.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
BFCL V4 (December 2025 archive) 37.3% ▁▁▁▂██●▅▆▂▂▂ beats 42/77 hard · avg 31% 3.0% at its level
Finance Agent Benchmark v1 (537 q… 3.1% ●·▃▃·█▅▅▅▅▃▅ beats 2/18 very hard · avg 5.5% 2.7% sharp
BFCL v3 52.9% ▁▁▁▁▁▃▅●▅▄▁▁ beats 49/114 medium · avg 51% 1.4% blunt · at its level
Multi-SWE-bench, 7 languages (Mag… 6.1% ▄···▄▄·▄▄·▄● beats 6/7 very hard · avg 2.6% 1.1% blunt · too hard for it
Galileo Agent Leaderboard v1 (too… 74.1% ▃▃●▃▆▅▆▅█▂▅▂ beats 4/31 easy · avg 78% 0.9% typical
coding · #221 of 469 · 17% of its score
Beats [Qwen3.5 4B](qwen3-5-4b.md), [Qwen3-VL 32B Instruct](qwen3-vl-32b-instruct.md), [Solar Open 100B (102B-A12B)](solar-open-100b.md) here, though they rank higher overall. Trails [QwQ-32B-Preview](qwq-32b-preview.md), [Ministral 3 3B Reasoning 2512](ministral-3-3b-reasoning.md), [s1.1 32B](s1-1-32b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Aider polyglot (225 exercises, tw… 15.6% ▂·▂●▅▂▃█▆▄▂▂ beats 6/49 very hard · avg 7.6% 13% sharp · at its level
WebApp1K-React pass@1 (onekq-ai) 76.9% ▃▁▁█▁▄▅●▄▇▁▄ beats 19/31 medium · avg 40% 2.5% sharp · at its level · noisy
Multi-SWE-bench, 7 languages (Mag… 6.1% ▄···▄▄·▄▄·▄● beats 6/7 very hard · avg 2.6% 1.1% blunt · too hard for it
chat & writing · #197 of 402 · 14% of its score
Beats [Gemma 2 9B IT SimPO](gemma-2-9b-simpo.md), [Granite 4.2 8B](granite-4-2-8b.md), [Mistral Small 4 (119B-A6.5B)](mistral-small-4.md) here, though they rank higher overall. Trails [LFM2.5 1.2B Instruct](lfm2-5-1-2b-instruct.md), [LFM2.5 1.2B Thinking](lfm2-5-1-2b-thinking.md), [LFM2.5 8B-A1B](lfm2-5-8b-a1b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Arena-Hard v2.0 hard prompts (sty… 17.2% ▂·▃▅●·▃█▆▂▂▂ beats 7/22 very hard · avg 12% 6.6% sharp · at its level
LMArena (text) 1327.0 ▁▁▁▂▃▃▃●▆▇█▃ beats 165/385 avg 1260 6.3% blunt
Arena-Hard v2.0 creative writing … 10.5% ●▂▂▆▂▂▃█▂▃·▂ beats 2/22 hard · avg 19% 1.5% sharp · at its level · noisy
math · #203 of 431 · 19% of its score
Beats [MiniMax-M2.1](minimax-m2-1.md), [DeepSeek V3 (Dec '24)](deepseek-v3.md), [Qwen3 235B-A22B (Apr 2025, hybrid)](qwen3-235b-a22b.md) here, though they rank higher overall. Trails [Mixtral 8x7B v0.1 (base)](mixtral-8x7b-v0-1.md), [Olmo 3 7B Instruct](olmo-3-7b-instruct.md), [Phi-4-mini-reasoning 3.8B](phi-4-mini-reasoning.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
MATH level 5 73.0% ▂▂▅▆▄█▆●▅▂▃▄ beats 62/92 hard · avg 28% 13% sharp · at its level
OTIS mock AIME 2024–2025 20.6% ▃▃▄▃●▃▇▇█▃▂▅ beats 58/197 very hard · avg 7.2% 4.3% sharp · at its level · noisy
MathNet-Solve-Test micro-average … 14.7% ●▃···█▃·▃·▃▃ beats 0/8 very hard · avg 15% 1.6% at its level
vision & documents · #101 of 153 · 11% of its score
Beats [DeepSeek V4 Flash Vision (Exp)](deepseek-v4-flash-vision-exp.md), [GPT-4o (2024-08-06)](gpt-4o-2024-08-06.md), [GPT-4o (2024-05-13)](gpt-4o-2024-05-13.md) here, though they rank higher overall.
test result vs the others in its run difficulty counts why
Arena vision (style controlled) 1145.4 ▁▁▂▂▂▂▃●▅▅█▇ beats 46/152 avg 1061 4.2% blunt
CharXiv (reasoning questions) 55.0% ▂▅▃▅▅●▆▅▆█▆▂ beats 35/101 medium · avg 36% 2.7% at its level · small run
CharXiv descriptive questions 87.7% ▂▄▄▅▇▄█●▇▂▁▂ beats 41/59 easy · avg 71% 2.2% small run
VMMU (Vietnamese multimodal under… 38.4% ▅······▃·●·█ beats 3/7 very hard · avg 36% 1.4% noisy
web search · #57 of 84 · 2.7% of its score
Beats [Kimi K2.6](kimi-k2-6.md), [GPT-5 Pro](gpt-5-pro.md), [DeepSeek V4 Flash (Preview, Apr 2026)](deepseek-v4-flash-preview.md) here, though they rank higher overall. Trails [GPT-4o mini (2024-07-18)](gpt-4o-mini-2024-07-18.md), [Mistral Small 3.1 24B (2503)](mistral-small-3-1.md), [Llama 4 Scout (109B-A17B)](llama-4-scout.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Finance Agent Benchmark v1 (537 q… 3.1% ●·▃▃·█▅▅▅▅▃▅ beats 2/18 very hard · avg 5.5% 2.7% sharp
other tests (10)
test result vs the others in its run difficulty counts why
GPQA Diamond 67.0% ▁▁▁▁▂▄▆▄●▇█▅ beats 101/211 hard · avg 39% 10% sharp · at its level
KCL-Essay vanilla 25.6% ▁···▁▃●█▂▃▃▂ beats 6/23 hard · avg 17% 4.0% at its level
EQ-Bench 3 (emotional intelligenc… 1062.0 ▁·▂▁●▅▅▃█▄▁▁ beats 8/62 avg 1034 3.8% typical
CPTUB average (SpeakLeash, 0-5 ju… 78.7% ▁▂▂▂▃█▃▆▅●▂▃ beats 77/95 medium · avg 65% 3.8% typical
KINA (Knowledge Index of Noah's A… 31.6% ▁▁▂▁▃▄█●▄▅▂▃ beats 24/39 very hard · avg 21% 3.4% typical
KCL-MCQA vanilla 43.3% ▂···█▂▆▆●█▆▂ beats 13/23 very hard · avg 29% 1.9% typical
LongSocialBench (full-context, 4-… 41.2% ▆▂·▄●·▂▂·▄▄▂ beats 6/16 very hard · avg 36% 1.8% blunt
DiagnosticIQ industrial maintenan… 63.3% ▁··▃▁▁▁▅●▅▁▅ beats 15/24 hard · avg 51% 1.3% blunt · at its level
Open PT-LLM Leaderboard average (… 80.1% ▃▁▁▁▁▁▁▁▂█●▁ beats 260/276 medium · avg 72% 1.2% at its level · noisy
Vectara hallucination leaderboard… 91.8% ▁▂▁▂▅█●▃▄▃▁▁ beats 62/98 very easy · avg 90% 0.0% blunt
source results share of the score
Epoch AI 3 27% ▇▇▇
Aider polyglot benchmark runs 1 13% ▇▇
LMArena leaderboard dataset 2 11% ▇
Arena-Hard-Auto v2.0-Preview leaderboard 2 8% ▇
Korean Canonical Legal Benchmark: Towar… 2 6% ▇
Apriel-1.5-15b-Thinker 2 5% ▇
Berkeley Function Calling Leaderboard r… 2 4% ▇
EQ-Bench 3 canonical leaderboard results 1 4% ▇
SpeakLeash CPTUB 1 4% ▇
Knowledge Index of Noah's Ark 1 3% ▇
Finance Agent Benchmark 1 3% ▇
WebApp1K models leaderboard 1 3% ▇
LongSocialBench: Do Long-Context LLMs U… 1 2% ▇
MathNet: a Global Multimodal Benchmark … 1 2% ▇
VMMU: A Vietnamese Multitask Multimodal… 1 1% ▇
DiagnosticIQ: A Benchmark for LLM-Based… 1 1% ▇
Open Portuguese LLM Leaderboard 1 1% ▇
Multi-SWE-bench experiments 1 1% ▇
Galileo Agent Leaderboard 1 1% ▇
Vectara hallucination leaderboard 1 0% ▇
prior: pull toward the middle 0% ▇
weighted average = 57.2
Licences: Epoch AI ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Aider polyglot benchmark runs ([Apache-2.0](../../licenses/aider-polyglot-runs-Apache-2.0.txt)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Arena-Hard-Auto v2.0-Preview leaderboard ([Apache-2.0](../../licenses/arena-hard-auto-Apache-2.0.txt)) · Korean Canonical Legal Benchmark: Toward Knowledge-Independent Evaluation of LLMs’ Legal Reasoning Capabilities ([CC BY 4.0](../../licenses/cards-final-NOTICE.txt)) · Apriel-1.5-15b-Thinker ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)) · Berkeley Function Calling Leaderboard result archive ([Apache-2.0](../../licenses/bfcl-result-archive-v4-Apache-2.0.txt)) · EQ-Bench 3 canonical leaderboard results ([MIT](../../licenses/eqbench3-MIT.txt)) · SpeakLeash CPTUB ([MIT](../../licenses/hf-leaderboard-spaces-MIT.txt)) · Knowledge Index of Noah's Ark ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Finance Agent Benchmark ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)) · WebApp1K models leaderboard ([MIT](../../licenses/webapp1k-leaderboard-MIT.txt)) · LongSocialBench: Do Long-Context LLMs Understand Online Discussion Threads? ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · MathNet: a Global Multimodal Benchmark for Mathematical Reasoning and Retrieval ([CC BY 4.0](../../licenses/paper-2604.18584v2-CC-BY-4.0.txt)) · VMMU: A Vietnamese Multitask Multimodal Understanding and Reasoning Benchmark ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)) · DiagnosticIQ: A Benchmark for LLM-Based Industrial Maintenance Action Recommendation from Symbolic Rules ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Open Portuguese LLM Leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt)) · Multi-SWE-bench experiments ([Apache-2.0](../../licenses/multi-swe-bench-experiments-Apache-2.0.txt)) · Galileo Agent Leaderboard ([Apache-2.0](../../licenses/galileo-agent-leaderboard-Apache-2.0.txt)) · Vectara hallucination leaderboard ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0))
## How to run it
### Through an API
api (Amazon Bedrock) · $0.42/M blended · $0.24 in / $0.97 out per 1M tokens · [docs.aws.amazon.com](https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-meta-llama-4-maverick-17b-instruct.html)
[price source](https://aws.amazon.com/bedrock/pricing): USD per 1M tokens, Amazon Bedrock on-demand Llama 4 Maverick 17B (Meta Llama 4 On-Demand table), US East (N. Virginia)/Ohio and US West (Oregon/N. California). Batch pricing excluded. Table cells are rendered from AWS's own price data file for the page (published 2026-09-25T06:55:02Z). Currency: USD. Amazon Bedrock official pricing page, observed 2026-09-25. · [context source](https://github.com/meta-llama/llama-models/blob/0e0b8c519242d5833d8c11bffc1232b77ad7f301/models/llama4/MODEL_CARD.md): Meta Llama 4 model card (llama-models repo, pinned): Llama 4 Maverick (17Bx128E) context length 1M (stated rounded). Model ceiling, not a promise by any hosted API. Observed 2026-09-25.
### 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 Llama 4 Maverick (402B-A17B) (ud-q2_k_xl, needs ~121.1 GB, from [huggingface.co/unsloth](https://huggingface.co/unsloth/Llama-4-Maverick-17B-128E-Instruct-GGUF/tree/main))
```sh
llama cli -hf unsloth/Llama-4-Maverick-17B-128E-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/Llama-4-Maverick-17B-128E-Instruct-GGUF:UD-Q2_K_XL
```
Other versions: ud-iq1_s (~60.8 GB) `llama cli -hf unsloth/Llama-4-Maverick-17B-128E-Instruct-GGUF:UD-IQ1_S`
### 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 Llama 4 Maverick (402B-A17B) (ud-q2_k_xl, needs ~121.1 GB, from [huggingface.co/unsloth](https://huggingface.co/unsloth/Llama-4-Maverick-17B-128E-Instruct-GGUF/tree/main))
```sh
lms get https://huggingface.co/unsloth/Llama-4-Maverick-17B-128E-Instruct-GGUF@UD-Q2_K_XL -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: ud-iq1_s (~60.8 GB) `lms get https://huggingface.co/unsloth/Llama-4-Maverick-17B-128E-Instruct-GGUF@UD-IQ1_S -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 Llama 4 Maverick (402B-A17B) (Q4_K_M, needs ~241.7 GB, from [ollama.com](https://ollama.com/library/llama4/tags))
```sh
ollama run llama4:maverick
```
Model repository and documentation: [huggingface.co/meta-llama](https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E-Instruct)
[interactive version](https://whatmodel.sh/model/llama-4-maverick) · [JSON](https://whatmodel.sh/api?format=json&all=1)