Llama 4 Maverick API Price Rises 23% — What It Costs You Now

The Numbers

Llama 4 Maverick's API pricing has moved up across the board. Input tokens rise from $0.188 → $0.20 per 1M tokens, and output tokens jump from $0.652 → $0.80 per 1M tokens — a blended increase of roughly +23%.

Output is where the pain lands hardest. At $0.80/1M, output tokens now cost about 4× the input rate, which matters most for workloads that generate long completions — summarisation, code generation, agentic loops.

Does It Actually Move the Needle?

For low-volume hobby projects, probably not. A developer pushing 5M output tokens a month goes from ~$3.26 to $4.00 — a $0.74 difference.

At scale it compounds fast. 500M output tokens/month shifts from ~$326 to $400 — an extra $74/month for nothing new in return. Pipelines running continuous inference should re-run their cost models now.

What to Do

• Audit your output-to-input ratio. Output-heavy workloads absorb this hike disproportionately.
• Compare alternatives. Several comparable open-weight models remain available at lower price points on our site. Check live rates before assuming Maverick is still the cheapest fit for your use case.
• Watch for provider variance. Not all API providers update simultaneously — shopping across hosts may recover some margin short-term.

For current pricing, benchmark scores, and a full price history, see Llama 4 Maverick — live specs & price history.

Ezra, Scout AI Team