Granite 3.3 8B (Non-reasoning)
Intelligence
7
#508/523
Coding
3.4
#396/429
Math
6.7
#242/265
Speed
416 tok/s
TTFT: 7.21s
Pricing
$0.03 / $0.25
per 1M tokens (in/out)
Granite 3.3 8B (Non-reasoning) is IBM’s model designed for various applications requiring natural language processing. It processes at 415.615 tokens per second and is priced at $0.03 per million input tokens, targeting professional users.
When to Use Granite 3.3 8B (Non-reasoning)
✓ Best For
- Text generation and completion tasks.
- Basic coding assistance.
- Mathematical problem-solving.
✗ Not Ideal For
- Complex reasoning tasks.
- High-speed applications requiring low latency.
How Granite 3.3 8B (Non-reasoning) Compares
Intelligence Index · Higher is better
IBMLiquid AIAlibaba
Benchmark Profile
Coding Index
AlibabaAllen Institute for AIIBMLG AI ResearchAI21 Labs
Output Speed · tok/s
InceptionIBMStepFunOpenAI
Math Index
AmazonMicrosoftIBM
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (30)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 7 | |
| Coding Index | 3.4 | |
| Math Index | 6.7 | |
| MMLU-Pro | 468% | |
| GPQA | 338% | |
| LiveCodeBench | 127% | |
| HLE | 42% | |
| SciCode | 10.1% | |
| IFBench | 22.4% | |
| LCR | 4.3% | |
| Tau2 | 10.5% | |
| AIME | 4.7% | |
| AIME 2025 | 6.7% | |
| MATH 500 | 66.5% | |
| AIME 2024 | 81.2% | |
| AlpacaEval 2.0 | 62.7% | |
| Arena Hard | 0.6 | |
| AttaQ | 88.5% | |
| BIG-Bench Hard | 69.1% | |
| DROP | 59.4% | |
| GSM8k | 80.9% | |
| HumanEval | 89.7% | |
| HumanEval+ | 86.1% | |
| IFEval | 74.8% | |
| MATH-500 | 69% | |
| MMLU | 65.5% | |
| PopQA | 26.2% | |
| TruthfulQA | 66.9% | |
| Arena Chat | 6.4 | |
| Arena Coding | -1.8 |
Data: Artificial Analysis · Updated: April 10, 2026
Frequently Asked Questions (15)
When was Granite 3.3 8B (Non-reasoning) released?
Granite 3.3 8B (Non-reasoning) was released on April 16, 2025.
Who created Granite 3.3 8B (Non-reasoning)?
Granite 3.3 8B (Non-reasoning) was created by IBM.
How intelligent is Granite 3.3 8B (Non-reasoning)?
Granite 3.3 8B (Non-reasoning) scores 7 on the Artificial Analysis Intelligence Index, placing it at the lower end among other open weight non-reasoning models of similar size (median: 11).
How fast is Granite 3.3 8B (Non-reasoning)?
Granite 3.3 8B (Non-reasoning) generates output at 319.9 tokens per second (based on the median across providers serving the model), which is well above average compared to other open weight non-reasoning models of similar size (median: 98.2 t/s).
What is the latency of Granite 3.3 8B (Non-reasoning)?
Granite 3.3 8B (Non-reasoning) has a time to first token (TTFT) of 21.60s (based on the median across providers serving the model), which is at the higher end compared to other open weight non-reasoning models of similar size (median: 1.73s).
How much does Granite 3.3 8B (Non-reasoning) cost?
Granite 3.3 8B (Non-reasoning) costs $0.03 per 1M input tokens (very competitive, median: $0.15) and $0.25 per 1M output tokens (better than average, median: $0.30), based on the median across providers serving the model.
What is Granite 3.3 8B (Non-reasoning) API pricing?
Granite 3.3 8B (Non-reasoning) costs $0.03 per 1M input tokens and $0.25 per 1M output tokens (based on the median across providers serving the model). For a blended rate (3:1 input to output ratio), this is $0.09 per 1M tokens. Pricing may vary by provider.
How verbose is Granite 3.3 8B (Non-reasoning)?
When evaluated on the Intelligence Index, Granite 3.3 8B (Non-reasoning) generated 8.3M output tokens, which is better than average compared to other open weight non-reasoning models of similar size (median: 8.5M).
Is Granite 3.3 8B (Non-reasoning) a reasoning model?
No, Granite 3.3 8B (Non-reasoning) is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
What input modalities does Granite 3.3 8B (Non-reasoning) support?
Granite 3.3 8B (Non-reasoning) supports text input.
What output modalities does Granite 3.3 8B (Non-reasoning) support?
Granite 3.3 8B (Non-reasoning) supports text output.
Can Granite 3.3 8B (Non-reasoning) process images?
No, Granite 3.3 8B (Non-reasoning) does not support image input. It can only process text.
Is Granite 3.3 8B (Non-reasoning) multimodal?
No, Granite 3.3 8B (Non-reasoning) is not multimodal. It only supports text input.
What is the context window of Granite 3.3 8B (Non-reasoning)?
Granite 3.3 8B (Non-reasoning) has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.
Is Granite 3.3 8B (Non-reasoning) open source?
Yes, Granite 3.3 8B (Non-reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
