Llama 4 Maverick 17B 128E Instruct FP8
watsonx/meta-llama/llama-4-maverick-17b-128e-instruct-fp8 · family llama · in: text, image → out: text · price provider-declared (via models.dev). Capabilities: tools text image.
$0.65
Blended $/1M (3:1)
$0.371
Input $/1M
$1.48
Output $/1M
$1.85
1M in + 1M out
—
Cache read $/1M
#3322 of 7,708
Cheapest rank
- Provider
- watsonx.ai (
watsonx) - Kind
- text (outputs text)
- Context
- 131,072 tokens
- Max output
- 8,192 tokens
- Capabilities
- tools text image
Same model, other providers
“Llama 4 Maverick 17B 128E Instruct FP8” is offered by 3 providers — cheapest-effective first. Azure is the cheapest at $0.44 blended.
| Provider | Input | Output | Blended | Context |
|---|---|---|---|---|
| Azure | $0.25 | $1 | $0.44 | 1M |
| Azure Cognitive Services | $0.25 | $1 | $0.44 | 1M |
| watsonx.ai (this) | $0.371 | $1.48 | $0.65 | 131k |
Call it
jishie indexes & prices models; it does not proxy inference.
Most providers are OpenAI-compatible — point base_url at watsonx.ai and pass this model id:
curl $BASE_URL/v1/chat/completions \
-H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" \
-d '{"model":"meta-llama/llama-4-maverick-17b-128e-instruct-fp8","messages":[{"role":"user","content":"hello"}]}'
Route programmatically with MCP find_model / get_model, or fetch this record free (no wallet):