workhorse
Grok 3 vs Llama 4 Maverick 17b 128e Instruct
Side-by-side pricing, context window, features, and per-use-case verdict. Data sourced from LiteLLM, updated regularly.
Specifications
| Feature | Grok 3 | Llama 4 Maverick 17b 128e Instruct |
|---|---|---|
| Provider | xai | groq |
| Model ID | xai/grok-3 | groq/meta-llama/llama-4-maverick-17b-128e-instruct |
| Input / 1M tokens | $3.00 | $0.20 (93% cheaper) |
| Output / 1M tokens | $15.00 | $0.60 (96% cheaper) |
| Cached input / 1M | $0.75 | — |
| Context window | 131kk tokens | 131kk tokens |
| Max output | 131kk tokens | 8kk tokens |
| Vision | ⬜ | ✅ |
| Function calling | ✅ | ✅ |
| Prompt caching | ✅ | ⬜ |
| Web search | ✅ | ⬜ |
| JSON schema | ⬜ | ✅ |
Monthly cost comparison
No cache discount applied — pure token costs at three usage scales.
| Scale | Grok 3 | Llama 4 Maverick 17b 128e Instruct | Winner |
|---|---|---|---|
| Small 1M in / 0.2M out | $6.00 | $0.32 | Llama 4 Maverick 17b 128e Instruct saves 95% |
| Medium 10M in / 2M out | $60.00 | $3.20 | Llama 4 Maverick 17b 128e Instruct saves 95% |
| Large 100M in / 20M out | $600 | $32.00 | Llama 4 Maverick 17b 128e Instruct saves 95% |
Category winners
Cheapest input
Llama 4 Maverick 17b 128e Instruct
$0.20/M
· 93% less than the other
Cheapest output
Llama 4 Maverick 17b 128e Instruct
$0.60/M
· 96% less than the other
Larger context
Grok 3
131kk tokens
More features
Grok 3
3 / 5 capabilities
(tied)
Better for…
High-volume RAG & repeated context
$0.75/M cached input — prompt caching makes repeated context retrieval significantly cheaper
Grok 3
Short interactions & high-throughput APIs
$0.60/M output — 96% cheaper on output at scale
Llama 4 Maverick 17b 128e Instruct
Cost-sensitive production at scale
$32.00/month at 100M-token scale — 95% cheaper than the alternative
Llama 4 Maverick 17b 128e Instruct
Deep-dive on each model
Pricing data sourced from LiteLLM and refreshed regularly. Always verify with each provider's official pricing page before making business decisions.