IBM, open weights
Granite 4.0 Micro
Granite 4.0 Micro by IBM ranks 318th of 354 ranked models on the Noometry Index as of October 2026, with a score of 29.0. Its strongest category is instruction following, where it ranks 169th. API pricing starts at $0.017 per million input tokens and $0.11 per million output tokens, with a 131K-token context window.
Last verified
Specifications
- Noometry rank
- #318 of 354
- Index score
- 29.0
- Evidence
- Confirmed 8 results
- Provider
- IBM
- Released
- October 2, 2025
- Weights
- Open weights
- Reasoning
- Unknown
- Context window
- 131K
- Max output
- 118K
- Input price
- $0.017 / M
- Output price
- $0.11 / M
- Blended price
- $0.0408 / M
- Output speed
- Not measured
- Value
- #3 of 219
- Knowledge cutoff
- Unknown
- Input
- text
- Hugging Face
- ibm-granite/granite-4.0-h-micro
Category scores
Each category score combines every public result we have in that category.
- Reasoning 19.2
- Math 12.0
- Knowledge 9.9
- Instruction Following 69.9
- Writing & Preference 46.7
| Category | Score | Rank | Results |
|---|---|---|---|
| Reasoning | 19.2 | #265 | 1 |
| Math | 12.0 | #307 | 2 |
| Knowledge | 9.9 | #304 | 3 |
| Instruction Following | 69.9 | #169 | 1 |
| Writing & Preference | 46.7 | #216 | 1 |
Strengths and weaknesses
Categories where Granite 4.0 Micro places highest and lowest among the models ranked in each, with its score against that category's median.
Strongest categories
| Category | Score | vs median | Rank |
|---|---|---|---|
| Instruction Following | 69.9 | −1.4 | #169 of 305, top 56% |
| Writing & Preference | 46.7 | −7.1 | #216 of 312, top 70% |
Closest competitors
The models ranked just above and below Granite 4.0 Micro. When scores are this close, price and speed are often the better way to choose.
| Model | Rank | Score | Blended $/M | Speed | |
|---|---|---|---|---|---|
| Phi 3 Small 8k Instruct | #314 | 29.3 | — | — | Compare |
| Claude 3.5 Haiku | #315 | 29.2 | — | — | Compare |
| GPT-4 | #316 | 29.1 | $37.50 | — | Compare |
| Llama 2-7B | #317 | 29.1 | — | — | Compare |
| Claude 3 Sonnet | #319 | 29.0 | — | — | Compare |
| Qwen2.5 7B Instruct | #320 | 29.0 | $0.31 | — | Compare |
| Llama 3.2 3B | #321 | 28.9 | $0.12 | — | Compare |
| Qwen1.5 4b Chat | #322 | 28.8 | — | — | Compare |
Sponsored placements are available on pages like this one. Advertise on Noometry
Benchmark results
Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.
Reasoning
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| Chess Puzzles | 0% | #117 of 129, top 91% | Epoch AI | 2026-08-30 |
Math
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.8% | #155 of 173, top 90% | Epoch AI | 2026-08-30 | |
| Omni-MATH | 20.9% | #51 of 57, top 90% | HELM Capabilities |
Knowledge
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| GPQA Diamond | 28.3% | #172 of 186, top 93% | Epoch AI | 2026-08-30 | |
| MMLU-Pro | 39.5% | #55 of 58, top 95% | HELM Capabilities | ||
| GPQA (HELM) | 30.7% | #53 of 57, top 93% | HELM Capabilities |
Instruction Following
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| IFEval | 84.9% | #20 of 57, top 36% | HELM Capabilities |
Writing & Preference
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| WildBench | 67% | #55 of 57, top 97% | HELM Capabilities |
API pricing by provider
| Route | Input $/M | Output $/M | Cached input $/M | Checked |
|---|---|---|---|---|
| openrouter | $0.017 | $0.11 | — | 2026-10-10 |
Compare Granite 4.0 Micro
- Granite 4.0 Micro vs Llama 2-7B
- Granite 4.0 Micro vs Claude 3 Sonnet
- Granite 4.0 Micro vs GPT-4
- Granite 4.0 Micro vs Qwen2.5 7B Instruct
- Granite 4.0 Micro vs Claude 3.5 Haiku
- Granite 4.0 Micro vs Llama 3.2 3B
- Granite 4.0 Micro vs GPT-6 Astra
- Granite 4.0 Micro vs Claude Fable 5.1
- Granite 4.0 Micro vs Gemini 3.8 Flash
- Granite 4.0 Micro vs Kimi K3
- Granite 4.0 Micro vs Grok 4.6
- Granite 4.0 Micro vs Qwen3.8 Max
- Granite 4.0 Micro vs GLM-5.3
- Granite 4.0 Micro vs Muse Spark 1.3
Other IBM models
Frequently asked questions
How good is Granite 4.0 Micro?
Granite 4.0 Micro by IBM ranks 318th of 354 ranked models on the Noometry Index as of October 2026, with a score of 29.0. Its strongest category is instruction following, where it ranks 169th. API pricing starts at $0.017 per million input tokens and $0.11 per million output tokens, with a 131K-token context window.
How much does Granite 4.0 Micro cost?
Granite 4.0 Micro costs $0.017 per million input tokens and $0.11 per million output tokens on openrouter.
What is Granite 4.0 Micro's context window?
Granite 4.0 Micro accepts up to 131K tokens of input and can write up to 118K tokens in one response.
Is Granite 4.0 Micro open source?
Yes. Granite 4.0 Micro's weights are downloadable from Hugging Face (ibm-granite/granite-4.0-h-micro); check the license for commercial terms.
What are Granite 4.0 Micro's strengths and weaknesses?
Relative to other ranked models, Granite 4.0 Micro places best in instruction following, writing & preference and lowest in knowledge, math.
What is Granite 4.0 Micro best at?
Its best category is instruction following, where it ranks 169th on Noometry.