Llama 4 Maverick

Open Source

Meta•Released on 2025-04-05

Meta's flagship open-source multimodal model. 17B active parameters with 400B total (128 expert MoE). 1M context window, natively multimodal with early fusion. Extremely cost-effective at $0.15/$0.60 per M tokens. Supports 12 languages.

80
Overall Score

Core Specs

1049K
Context Window
16K
Max Output
✗
Reasoning
✓
Open Source
Multimodal Support
textimage

Scenario Scores

Coding
#20 / 2478
Writing
#14 / 2380
Translation
#14 / 2382
Data Analysis
#18 / 2479
Conversation
#14 / 2383
Image Gen
—

User Feedback Highlights

Based on community feedback. Hover to see original reviews.

− Knowledge cutoff August 2024− Benchmark gaming controversy+ Extremely affordable ($0.15/$0.60)+ 1M context window− 16K max output limit+ Native multimodal (text + image)− Coding performance below Claude/GPT+ Open source (Llama 4 Community License)+ High throughput MoE architecture
Sentiment:👍 65%😐 20%👎 15%

Pros & Cons

Pros

  • +Extremely affordable ($0.15/$0.60)
  • +1M context window
  • +Native multimodal (text + image)
  • +Open source (Llama 4 Community License)
  • +High throughput MoE architecture

Cons

  • −Coding performance below Claude/GPT
  • −Benchmark gaming controversy
  • −16K max output limit
  • −Knowledge cutoff August 2024

Reliability

Incidents (30d)0
开源模型，可靠性取决于托管方

Pricing

Input (per 1M tokens)$0.15
Output (per 1M tokens)$0.60
Free trial available
Updated on 2026-03-07

Compare with Others

Llama 4 Maverick vs GPT-5.4 →Llama 4 Maverick vs GPT-5.4 Thinking →Llama 4 Maverick vs Claude Opus 4.6 →

Benchmarks

mmmu73.4%
codingAccuracy70%
lmArenaElo1417%