Grok 4.3 vs Llama 4 Maverick: cost per task
Llama 4 Maverick is cheaper on 12 of 14 tasks and about 3.5x cheaper on average (geometric mean). Grok 4.3 has the higher LMArena Text score (1442 vs 1327).
| Task | Grok 4.3 | Llama 4 Maverick | Difference |
|---|---|---|---|
| Answer a chat question | 0.45¢ | 0.11¢ | 4.3x |
| Reply to a customer email | 0.27¢ | 0.05¢ | 5.3x |
| Write a 1,500-word blog post | 1.00¢ | 1.0¢ | 1.0x |
| Write 10 product descriptions | 0.43¢ | 1.1¢ | 2.4x |
| Translate 1,000 words | 1.3¢ | 0.11¢ | 11.9x |
| Proofread one page (500 words) | 0.60¢ | 0.13¢ | 4.5x |
| Summarize a 20-page PDF | 2.0¢ | 0.42¢ | 4.8x |
| Extract data from an invoice | 0.55¢ | 0.15¢ | 3.6x |
| Analyze a 200-row spreadsheet | 8.6¢ | 1.2¢ | 7.4x |
| Answer from 5 retrieved documents (RAG) | 0.89¢ | 0.26¢ | 3.5x |
| Classify 100 support tickets | 1.0¢ | 0.58¢ | 1.8x |
| Review 500 lines of code | 9.5¢ | 1.1¢ | 8.3x |
| Write a SQL query | 0.55¢ | 0.17¢ | 3.2x |
| Run a 50-step coding agent | 73¢ | 19¢ | 3.9x |
Specs and list prices
| Grok 4.3 | Llama 4 Maverick | |
|---|---|---|
| Input per 1M | $1.25 | $0.19 |
| Output per 1M | $2.50 | $0.65 |
| Cached input per 1M | $0.20 | $0.05 |
| Context | 1M | 1M |
| Providers | 4 | 4 |
| Reasoning | default | none |
| LMArena Text | 1442 | 1327 |
| Released | May 1, 2026 | Apr 5, 2025 |
FAQ
Is Grok 4.3 or Llama 4 Maverick cheaper?
Llama 4 Maverick is cheaper on 12 of 14 tasks and about 3.5x cheaper on average (geometric mean). Grok 4.3 has the higher LMArena Text score (1442 vs 1327).
How much does Grok 4.3 cost compared to Llama 4 Maverick per token?
Grok 4.3: $1.25 input and $2.50 output per 1M tokens. Llama 4 Maverick: $0.19 input and $0.65 output. Token prices alone mislead - models use different numbers of tokens for the same task, which is what the task table shows.