Granite 3.3 8B (Non-reasoning)
Analysis Summary
Granite 3.3 8B (Non-reasoning) is IBM's small-scale language model, positioned for cost-sensitive deployments rather than complex reasoning or coding tasks. Its intelligence and coding benchmarks are low relative to the broader field, and its agentic score is minimal, making it unsuitable for autonomous or multi-step workflows.
For businesses, it fits narrow, well-defined tasks: basic content drafting, simple classification, or lightweight instruction-following where cost control matters more than output quality. Its instruction-following score (ifbench) is weak, and long-context reliability is poor, so it should not be used for document analysis or client-facing content without human review.
At $0.03 input and $0.25 output per million tokens, pricing is competitive for a small model, but the performance ceiling is low. Teams needing a cheap edge-deployable model for simple tasks may find it useful; those needing reliable business output should look to larger alternatives.
Assessed June 6, 2026
Editorial notes
Granite 3.3 8B is IBM's compact non-reasoning model with limited benchmark performance across coding and reasoning tasks, suited to lightweight instruction-following and low-cost content workflows.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How Granite 3.3 8B (Non-reasoning) compares
Granite 3.3 8B (Non-reasoning) ranks #364 of 378 AI models we track for overall intelligence, #290 of 315 for coding, #254 of 289 for agentic tasks. At $0.03 per million input tokens it is cheaper than 76% of comparable models.
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does Granite 3.3 8B (Non-reasoning) stack up?
Compare side-by-side with other professional models.
Model Information
| Provider | IBM |
| Release Date | April 16, 2025 |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.03 | $0.000030 |
| Output | $0.25 | $0.000250 |
Leaderboard Categories
Frequently asked questions about Granite 3.3 8B (Non-reasoning)
How much does Granite 3.3 8B (Non-reasoning) cost?
Granite 3.3 8B (Non-reasoning) costs $0.03 per million input tokens and $0.25 per million output tokens.
Is Granite 3.3 8B (Non-reasoning) good for coding?
On our coding benchmark index, Granite 3.3 8B (Non-reasoning) ranks #290 of 315 models, placing it in the broader range of the field for code generation and debugging.
Who created Granite 3.3 8B (Non-reasoning)?
Granite 3.3 8B (Non-reasoning) is developed by IBM and was released on April 16, 2025.
