Claude Fable 5$22.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/MGemini 1.5 Pro$2.375/MGemini 1.0 Ultra$12.000/MGemini 1.0 Pro$0.800/MClaude Fable 5$22.000/MClaude Opus 4.8$11.000/MClaude Opus 4.7$11.000/MClaude Opus 4.6$11.000/MClaude Opus 4.5$33.000/MClaude Sonnet 3.7$6.600/MClaude Opus 3$33.000/MClaude 2.1$12.800/MClaude 2$12.800/MGPT-5.5$12.500/MGPT-5.2$5.425/MGPT-5.2-Codex$5.425/MGPT-5$3.875/MGPT-4.5$97.500/MGPT-4 Turbo Preview$16.000/MGPT-4$39.000/MGPT-4-32k$78.000/Mo3$19.000/Mo3-mini$2.090/Mo4-mini$2.090/Mo1$28.500/Mo1-mini$5.700/Mo1-preview$28.500/MGemini 3.5 Pro$5.000/MGemini 3.1 Pro$5.000/MGemini 3 Pro$5.000/MGemini 2.5 Pro$3.875/MGemini 1.5 Pro$2.375/MGemini 1.0 Ultra$12.000/MGemini 1.0 Pro$0.800/M
BETA
Subscribe
← Back to Index·All IBM models
IBM
IBMEfficient

Granite 3.0 8B Dense

Previous Gen

Dense 8B Granite 3.0 — superseded by 3.1 but kept for legacy watsonx pipelines.

Granite 3.0 8B Dense is a efficient AI model from IBM. It costs $0.050 per million input tokens and $0.250 per million output tokens (blended $0.110/M), with a 4K-token context window.

Released Oct 2024Modalities textOfficial model page ↗API docs ↗Compare with another model →Estimate monthly cost →
INPUT
$0.050/M
per million input tokens
OUTPUT
$0.250/M
per million output tokens
BLENDED 70/30
$0.110/M
default reference rate · how it's calculated →
CONTEXT
4K
4,000 tokens
What it's good at
  • Apache 2.0
  • IBM indemnity
Typical use cases
  • Legacy Granite 3.0 deployments
Benchmarks
vs. best public score
MMLU66%
Multitask academic knowledge across 57 subjects.
GPQA Diamond34%
Graduate-level science questions, "Google-proof".
MATH31%
High-school competition math problems.
HumanEval65%
Python function synthesis from docstrings.
Hand-curated from each provider's published reports and public leaderboards. Methodology varies across sources — treat as directional rather than authoritative.
More from IBM
See all 5 →
Granite 3.1 8B Instruct
Compact · 128K ctx
in $0.050/Mout $0.250/M
Granite 3.1 2B Instruct
Nano · 128K ctx
in $0.030/Mout $0.100/M
Granite Code 34B
Code · 8K ctx
in $0.350/Mout $1.050/M
Granite 20B Multilingual
Multilingual · 8K ctx
in $0.400/Mout $0.400/M
Frequently asked questions

How much does Granite 3.0 8B Dense cost?

Granite 3.0 8B Dense costs $0.050 per million input tokens and $0.250 per million output tokens, for a blended reference rate of $0.110 per million tokens.

What is Granite 3.0 8B Dense's context window?

Granite 3.0 8B Dense supports up to 4K tokens of context (4,000 tokens).

What is Granite 3.0 8B Dense best for?

Granite 3.0 8B Dense is well suited to Apache 2.0, IBM indemnity and Legacy Granite 3.0 deployments.

Who makes Granite 3.0 8B Dense?

Granite 3.0 8B Dense is developed and served by IBM. It was released in Oct 2024.