impulse

Granite 3.3 8B Instruct

Open weights
API id granite-3-3-8b-instruct
Input / 1M
$0.20
at IBM watsonx.ai
Output / 1M
$0.20
at IBM watsonx.ai
Context window
8K
tokens
Max output
8K
tokens per reply
Hosts
2
2 listings
Released
-
date not published

Pricing

List prices at IBM watsonx.ai, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.20$0.20

What it costs a month

At IBM watsonx.ai list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out$75.00
  • RAG over documents2,000 requests a day, 8K in / 0.6K out$103
  • Coding agent1,000 requests a day, 40K in / 2K out$252
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$528
Use your own numbers

Where to run it

2 hosts serve Granite 3.3 8B Instruct. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
granite-3-3-8b-instruct
$0.20 / $0.20
ReplicateInference
granite-3.3-8b-instruct
$0.03 / $0.25
−58% vs headline

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text
Features
Parallel tool calls
Open weights
License: Apache-2.0
Runtimes: Ollama, llama.cpp, vLLM

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

  • Replicate
    granite-3.3-8b-instruct
    $0.03 / $0.25
    58% less

Smaller IBM models

No cheaper IBM model keeps its key capabilities.

Other frontier labs

No cheaper match from another frontier lab.

Open-weight picks

  • Qwen3.8 Flash
    Alibaba Qwen · on AIHubMix
    $0.113 / $0.38
    Has every key capability
    10% less
  • Nemotron 3.5 Super VL
    NVIDIA · on Kyma API
    $0.081 / $0.472
    Has every key capability
    11% less
  • Gemma 4 31B IT
    Google · on W&B Inference
    $0.10 / $0.34
    Has every key capability
    20% less
  • Ministral 8B 2512
    Mistral AI · on Mistral AI
    $0.15 / $0.15
    Has every key capability
    25% less

Model line

Newer and older versions, at each lab's own price.

  1. UndatedGranite 3.3 8B InstructLatest$0.20 / $0.20
  2. UndatedGranite 3.2 8B Instruct$0.03 / $0.25
  3. UndatedGranite 3.1 8B Instruct$0.03 / $0.25
  4. UndatedGranite 3 8B Instruct$0.20 / $0.20
  5. UndatedGranite 3.0 8B Instruct$0.05 / $0.25

Call it

Using IBM watsonx.ai's API.

Python
from litellm import completion

response = completion(
    model="watsonx/granite-3-3-8b-instruct",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Granite 3.3 8B Instruct cost?

On IBM watsonx.ai, Granite 3.3 8B Instruct costs $0.20 per million input tokens and $0.20 per million output tokens. The cheapest host is Replicate, at $0.03 per million input tokens and $0.25 per million output tokens.

Which APIs offer Granite 3.3 8B Instruct?

Granite 3.3 8B Instruct runs on 2 APIs: IBM watsonx.ai and Replicate.

What is Granite 3.3 8B Instruct's context window?

8K tokens, with up to 8K tokens in one reply.

Is Granite 3.3 8B Instruct being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of Granite 3.3 8B Instruct?

No. Granite 3.3 8B Instruct is the newest model in its line.

What are cheaper alternatives to Granite 3.3 8B Instruct?

Models that cost less and keep most of what it can do: Qwen3.8 Flash on AIHubMix (10% less). Granite 3.3 8B Instruct itself costs 58% less on Replicate. Savings are on a mix of 3 input tokens to 1 output token.

What is Granite 3.3 8B Instruct's API model id?

Use granite-3-3-8b-instruct with IBM watsonx.ai. Other hosts use their own ids: granite-3.3-8b-instruct on Replicate.

Sources

Raw catalog entry for granite-3-3-8b-instruct on IBM watsonx.ai
{
  "max_input_tokens": 8192,
  "max_output_tokens": 8192,
  "input_cost_per_token": 2e-7,
  "output_cost_per_token": 2e-7,
  "litellm_provider": "watsonx",
  "mode": "chat",
  "supports_function_calling": true,
  "supports_parallel_function_calling": true,
  "supports_vision": false
}