---
id: "ibm-granite/granite-4.1-8b"
also: ["ibm-granite/granite-4.1-8b-20260429", "ibm-granite/granite-4.1-8b"]
cheapest_list_usd_per_m: "0.05 in, 0.1 out, at coreweave/bf16 on openrouter"
cheapest_effective_usd_per_m: "0.0496 in, 0.0985 out, at coreweave/bf16 on openrouter"
context: 131072
max_output: 117964
input: ["text"]
output: ["text"]
released: "2026-04-30"
params: ["frequency_penalty", "logprobs", "max_tokens", "presence_penalty", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 1
gateways: "openrouter (1)"
no_training: 1
trains: 0
retains_prompts: 0
coding: 9.5000
---
# IBM: Granite 4.1 8B

Granite 4.1 8B is a dense, decoder-only 8-billion-parameter language model from IBM, part of the Granite 4.1 family. It supports a 131K-token context window and is designed for enterprise tasks...