69.4%
GPQA Diamond
Models / Meta
Llama 4 Maverick
Llama 4 · Released 2026-04-05
Meta's reasoning-focused Llama 4 MoE. 1M context, 17B active / 400B total params, 128 experts. Open weights for multimodal assistants and coding.
Default Llama 4 pick for quality-first self-hosting when Scout's 10M context isn't required.
Composite33
Context1M
Input / 1M$0.17
Output / 1M$0.60
Knowledge cutoff2024-08-01
Output speed50 t/s
Statusflagship
Family profile
Best published score in each covered benchmark family.
40/100 avg
- Reasoning
- 69
- Coding
- 32
- Agentic
- 8
- Long context
- 50
4 tested benchmark families
Benchmark placements
Per-benchmark positions
Where Llama 4 Maverick places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| SciCode | Coding | #92/ 97 | 32 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #121/ 195 | Artificial Analysis | 2026-10-12 | |
| AA-LCR | Long context | #159/ 184 | 50 | Artificial Analysis | 2026-09-28 |
| Terminal-Bench | Agentic | #159/ 167 | 8 | Artificial Analysis | 2026-09-28 |
| GPQA Diamond | Reasoning | #164/ 184 | 69 | Artificial Analysis GPQA Diamond evaluation | 2026-06-09 |
| Humanity's Last Exam | Reasoning | #178/ 186 | 5 | Artificial Analysis | 2026-09-28 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where Llama 4 Maverick shows up, it's highlighted.
ReasoningLlama 4 Maverick · 69
- 196
- 295
- 395
Math
- 152
- 252
- 350
CodingLlama 4 Maverick · 32
- 174
- 273
- 370
AgenticLlama 4 Maverick · 8
- 191
- 289
- 389
Long contextLlama 4 Maverick · 50
- 189
- 286
- 385
Newest receipts
5 recentLast 120 days
Benchmark rows added to the public ledger for Llama 4 Maverick in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 50 t/s | Artificial Analysis | 10 |
| AA-LCR | Long context | 50% | Artificial Analysis | 24 |
| Humanity's Last Exam | Reasoning | 4.9% | Artificial Analysis | 24 |
| SciCode | Coding | 31.7% | Artificial Analysis | 24 |
| Terminal-Bench | Agentic | 7.87% | Artificial Analysis | 24 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where Llama 4 Maverick sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIGoogleAlibabaMetaZhipuxAIDeepSeekMiniMaxXiaomiMoonshotOtherMeituanMistralNVIDIACohere
Full benchmark ledger
Every catalog benchmark for Llama 4 Maverick. Scores link to the original source; gaps mean no public row exists yet.
6 of 13 catalog benchmarks have a sourced row for Llama 4 Maverick.
Reasoning
Humanity's Last ExamVerified
4.9%
HLE (AA run)
Coding
SciCodeVerified
31.7%
SciCode (AA run)
Agentic
Terminal-BenchVerified
7.87%
Terminal-Bench v2.1 (AA run)
Long context
AA-LCRVerified
50%
AA-LCR (AA run)
Performance
AA output speedVerified
50 t/s
Median output tokens/s (1k prompt, default provider)
Sources
Where the numbers come from
- Meta Llama 4 blog· checked 2026-06-09
- Llama 4 model card· checked 2026-06-09
Questions this model page answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What is Llama 4 Maverick?
Llama 4 Maverick is a Meta model: Meta's reasoning-focused Llama 4 MoE. 1M context, 17B active / 400B total params, 128 experts. Open weights for multimodal assistants and coding.
What is Llama 4 Maverick for?
Default Llama 4 pick for quality-first self-hosting when Scout's 10M context isn't required.
What does Llama 4 Maverick cost?
Llama 4 Maverick lists $0.17 input and $0.60 output per 1M tokens. Pricing last checked 2026-06-09.
How large is the Llama 4 Maverick context window?
Llama 4 Maverick publishes a 1M token context window.
What is the VerdictPal status of Llama 4 Maverick?
Llama 4 Maverick carries Flagship status on the model atlas.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.