Llama 4 Maverick
Llama 4 MaverickAlternatives?
CheaperGPT-6 Lunamedium+61%-54%$0.017Function calling only with none · Chat Completions+3
OpenGLM 5.3max+86%+5,137%$2.01Memory (4-bit): 241 GB → 452 GB+5
CheaperClaude Haiku 5.5low+55%-36%$0.024Remove sampling parameters+5
CheaperAgnes 2.5 Pro Beta+54%-51%$0.10 / $0.30CheaperGPT-5.6 Lunamedium+47%-59%$0.016GPT-6 Lunamedium+61%-54%$0.017Function calling only with none · Chat Completions+3Claude Haiku 5.5low+55%-36%$0.024Remove sampling parameters+5Agnes 2.5 Pro Beta+54%-51%$0.10 / $0.30GPT-5.6 Lunamedium+47%-59%$0.016
Gemma 4 31BthinkingOpen+43%-47%$0.10 / $0.34Context: 1M → 256KGPT-6 Lunalow+48%-88%$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow+45%-74%$0.01GPT OSS 20BhighOpen+23%-70%$0.011Context: 1M → 131K+1
Mistral Small 4thinkingOpen+22%-61%$0.015Context: 1M → 262K
MiMo v2.5thinkingOpen+38%-50%$0.019GLM 5.3maxOpen+86%+5,137%$2.01Memory (4-bit): 241 GB → 452 GB+5
Step 5 PreviewthinkingOpen+86%+2,580%$1.03Memory (4-bit): 241 GB → 363 GB+3Muse Spark OpenxhighOpen+84%+2,102%$0.84Memory (4-bit): 241 GB → 493 GB+2
DeepSeek V4.1 Flash 1017maxOpen+76%+584%$0.26Memory (4-bit): 241 GB → 458 GB+5
Ling 3.1 FlashOpen+70%+2,510%$1.00Memory (4-bit): 241 GB → 343 GB+1Muse Wonder Full+97%——Only in Muse CodeMuse Wondermax+90%+11,690%$4.52Only auto tool_choice is supported+5Muse Spark 1.3xhigh+88%+3,471%$1.37Only auto tool_choice is supported+5Muse Spark 1.1xhigh+85%+3,504%$1.38Only n=1 · Meta / Chat Completions+1Muse Spark OpenxhighOpen+84%+2,102%$0.84Memory (4-bit): 241 GB → 493 GB+2Muse Spark 1.2xhigh+81%+2,445%$0.97Only auto tool_choice is supported+1Muse Glimmer 30BhighOpen+42%+50%$0.058Context: 1M → 131KLlama 4 Scoutnon-thinkingOpen-33%+193%$0.11All Meta models →Run it yourself
estimateFAQ
How much does Llama 4 Maverick cost?›
$0.19 per 1M input tokens and $0.65 per 1M output tokens (cheapest host: DigitalOcean). A typical benchmark task costs about $0.038 at default effort.
Is Llama 4 Maverick free to use?›
There is no free API; the cheapest paid price is $0.19 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Llama 4 Maverick, and what is it best at?›
#277 of 334 in Chat & Knowledge, #75 of 132 in Coding, #128 of 133 in Agents & Tool Use, #127 of 195 in Writing & Language, #90 of 124 in Vision & Documents.
Is Llama 4 Maverick better than Nemotron 3.5 Lightning?›
On SciCode, Llama 4 Maverick scores 31.7 vs 32.1 — about the same. It is 52% cheaper per task.
How can I use Llama 4 Maverick?›
Through the API of DigitalOcean, Google, Parasail.
What is the context window of Llama 4 Maverick?›
1M tokens, with up to 16K output tokens.
When was Llama 4 Maverick released? Is there a newer version?›
Released 2025-04-05. It is the current version in its line.
Is Llama 4 Maverick open source?›
Yes, the weights are open (llama4 licence).
Can I run Llama 4 Maverick locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Llama 4 Maverick read images?›
It accepts text and images as input.
What is Llama 4 Maverick?›
Llama 4 Maverick is an open-weights language model from Meta, released Apr 2025.
What are the best alternatives to Llama 4 Maverick?›
GPT-6 Luna (cheaper), GLM 5.3 (open weights).