Fireworks AI · none caching
Llama 3.2 11B Vision Instruct
Llama 3.2 11B Vision Instruct on Zumik: live pricing, context, and caching, routable by id or alias through one OpenAI-compatible endpoint.
Specifications
At a glance.
| Provider | Fireworks AI |
| Family | mllama |
| Released | — |
| License | Open weights |
| Context window | 131K tokens |
| Max output | — |
| Parameters | 11B |
| Modalities | text, image |
| Tool calling | No |
| Reasoning mode | No |
| Caching | none |
| Batch discount | No batch tier |
Measured by Zumik
What reuse looks like here.
Pricing, context, and capabilities for Llama 3.2 11B Vision Instruct are live, but it is outside the flagship set Zumik benchmarks in depth, so measured reuse, capture, and warm TTFT are not shown yet. Run a workload estimate or route it by id to start collecting traces.
Call it
Same OpenAI client, this model.
from openai import OpenAI
client = OpenAI(base_url="https://api.zumik.ai/v1", api_key="zk_live_...")
r = client.responses.create(
model="llama-v3p2-11b-vision-instruct",
input="Draft a fix for the failing test.",
)
print(r.usage.input_tokens_cached) # confirm reuseFrequently asked
Llama 3.2 11B Vision Instruct, answered.
How much does Llama 3.2 11B Vision Instruct cost?
Llama 3.2 11B Vision Instruct is an open-weights model routed through Fireworks AI. It is priced on the host's serverless size tier rather than a single published per-token list price, so it shows "—" here until profiled.
What is Llama 3.2 11B Vision Instruct's context window?
Llama 3.2 11B Vision Instruct supports a 131K-token context window.
Does Llama 3.2 11B Vision Instruct support prompt caching?
Yes. Fireworks AI uses Automatic prompt caching (serverless and dedicated) caching. In the Zumik corpus, Llama 3.2 11B Vision Instruct shows a median cache capture of 77% on agent workloads.
Run Llama 3.2 11B Vision Instruct with reuse measured.
Point an OpenAI client at Zumik and see exactly how much of this model's input you are reusing.