Llama 4 Maverick 17B Instruct
Llama 4 Maverick 17B Instruct is Meta's general-purpose model: 1M token context window, 5.24% on SWE-Bench Pro. It supports native tool calling, image/file input, open weights. Released 2025-04-05.
Context Window
1M
Max Output Tokens
16K
SWE-Bench Pro Score
5.24%
Tool Calling
Yes
All metrics
| Metric | Value | Confidence | Verified | Provenance receipt |
|---|---|---|---|---|
| Context Window | 1M | Verified · 2+ sources | Jun 10, 2026 | Receipt |
| Max Output Tokens | 16K | Verified · 2+ sources | Jun 10, 2026 | Receipt |
| SWE-Bench Pro Score | 5.24% | Verified · 2+ sources | Jun 10, 2026 | Receipt |
| Tool Calling | Yes | Verified · 2+ sources | Jun 10, 2026 | Receipt |
| Extended Reasoning | No | Verified · 2+ sources | Jun 10, 2026 | Receipt |
| Tokens per Wh (efficiency) | 7,435 | Single source | Jul 19, 2026 | Receipt |
Data sources
Facts on this page are sourced from the following verified sources.
- APImodels.devLast synced 3d ago
- third_partyML.ENERGY LeaderboardLast synced 3d ago
Compare Llama 4 Maverick 17B Instruct
Related models
#1DeepSeek V4 Flash
DeepSeek
DeepSeek V4 Flash is DeepSeek's reasoning-capable model: $0.14/$0.28 per 1M input/output tokens, 1M token context window.
#7Gemini 3.5 Flash
Gemini 3.5 Flash is Google's reasoning-capable model: $1.50/$9 per 1M input/output tokens, 1.0M token context window.
#9Claude Haiku 4.5 (latest)
Anthropic
Claude Haiku 4.5 (latest) is Anthropic's reasoning-capable model: $1/$5 per 1M input/output tokens, 200K token context window, 39.45% on SWE-Bench Pro.
#17o4-mini
OpenAI
o4-mini is OpenAI's reasoning-capable model: $1.10/$4.40 per 1M input/output tokens, 200K token context window.
Spotted an error? Suggest a correction.