llm11

Llama 3.2 3B Instruct with Jev

Llama 3.2 3B Instruct is a model from Meta, released 2024-09-25. It costs $0.050 per million input tokens and $0.330 per million output tokens, reads up to 131K tokens of context, and writes up to 117,964 tokens in one reply. Through llm11, set model to meta-llama/llama-3.2-3b-instruct.

Input price
$0.050 / 1M
Output price
$0.330 / 1M
Context window
131K tokens
Max output
118K tokens
Released
2024-09-25
Accepts
text

Read from the live catalogue and refreshed daily. At blended list price, 89% of the 332 priced models we can call cost more.

Using Llama 3.2 3B Instruct with Jev

Jev is TypeSafe AI's System One model. It reads each request first and decides which model in your pool answers, with a calibrated confidence on that call. Llama 3.2 3B Instruct does the answering when Jev sends it work, or when you name it yourself. At list price Llama 3.2 3B Instruct falls in the llm11-fast band, but no pack routes to it because Meta is not one of the labs the packs draw from. You can still pin it: pass meta-llama/llama-3.2-3b-instruct as model and routing steps aside while verification still runs.

Jev is new, so we have not published results for Llama 3.2 3B Instruct with and without it, and this page does not invent any. Every response comes with a receipt that names the model that answered and what it cost against your most expensive model, so you can measure the pairing on your own traffic.

About JevThe System One router

What a request costs

List price arithmetic, the same sum a receipt does. llm11 adds nothing per request; the only fee is on buying credits.

RequestTokens in / outOneA thousand
Short chat turn500 / 200$0.00009$0.091
Question over retrieved documents4,000 / 500$0.00036$0.365
Long document summary30,000 / 1,000$0.00183$1.83

Cheaper models to try

The catalogue has no benchmark score for Llama 3.2 3B Instruct, so this is a price ordering. It does not claim these answer as well.

Cheaper alternatives to Llama 3.2 3B Instruct
ModelInput / 1MOutput / 1MContext
Qwen Qwen3.5-Flash$0.065$0.2601M
Qwen Qwen3.5-9B$0.100$0.150262K
Mistral Ministral 3 3B 2512$0.100$0.100131K
DeepSeek DeepSeek V4 Flash 0731$0.018$0.3201.31M

Call it

Same OpenAI request shape. Naming the model pins it, so nothing is routed, and the answer is still checked.

python

from openai import OpenAI

client = OpenAI(base_url="https://www.llm11.com/v1", api_key="llm11_live_...")

res = client.chat.completions.create(
    model="meta-llama/llama-3.2-3b-instruct",
    messages=[{"role": "user", "content": "Hello"}],
)

More from Meta

Other Meta models
ModelInput / 1MOutput / 1MContext
Llama Guard 4 12B$0.180$0.180164K
Llama 4 Maverick$0.188$0.6521.05M
Llama 4 Scout$0.100$0.3001.31M
Llama 3.3 70B Instruct$0.100$0.320131K
Llama 3.2 1B Instruct$0.027$0.20160K
Llama 3.1 70B Instruct$0.400$0.400131K

All 8 Meta models

Questions

What is Llama 3.2 3B Instruct with Jev?
Jev is TypeSafe AI's System One model. It reads each request first and decides which model in your pool answers, with a calibrated confidence on that call. Llama 3.2 3B Instruct does the answering when Jev sends it work, or when you name it yourself. At list price Llama 3.2 3B Instruct falls in the llm11-fast band, but no pack routes to it because Meta is not one of the labs the packs draw from. You can still pin it: pass meta-llama/llama-3.2-3b-instruct as model and routing steps aside while verification still runs. Routing decisions are made by Jev, TypeSafe AI's System One model.
Is Llama 3.2 3B Instruct a System One model?
No. Llama 3.2 3B Instruct answers requests. The System One model is Jev, which decides which model answers each one, so the two do different jobs and llm11 uses both.
How much does Llama 3.2 3B Instruct cost?
$0.050 per million input tokens and $0.330 per million output tokens. A 4,000 token prompt with a 500 token answer costs about $0.00036, so a thousand of them is about $0.365. llm11 passes provider list price through and charges 5% when you buy credits.
What is the Llama 3.2 3B Instruct context window?
131,072 tokens, with up to 117,964 tokens in a single reply.
Does Llama 3.2 3B Instruct support tool calling and structured output?
The catalogue lists structured output, but it does not list tool calling and reasoning controls.
Can I use Llama 3.2 3B Instruct through llm11?
Yes. Set model to meta-llama/llama-3.2-3b-instruct on the OpenAI-compatible endpoint and the request goes to it directly, with verification still running.
Does Jev route to Llama 3.2 3B Instruct?
Routing decisions are made by Jev, TypeSafe AI's System One model. At list price Llama 3.2 3B Instruct falls in the llm11-fast band, but no pack routes to it because Meta is not one of the labs the packs draw from. You can still pin it: pass meta-llama/llama-3.2-3b-instruct as model and routing steps aside while verification still runs.
What is a cheaper alternative to Llama 3.2 3B Instruct?
Nearest in price and below it: Qwen Qwen3.5-Flash at $0.065 in and $0.260 out, Qwen Qwen3.5-9B at $0.100 in and $0.150 out and Mistral Ministral 3 3B 2512 at $0.100 in and $0.100 out. The catalogue has no benchmark score for Llama 3.2 3B Instruct, so this is a price ordering. It does not claim these answer as well.