Granite-4.0-H-Small
Stored tensors can include buffers and repeated shared weights. A verified unique parameter count is unavailable. Checkpoint size is not the memory required to run the model.
GeekRadar / The comparison desk
COMPARE.
AI modelsGranite-4.0-H-Small
AI model Released
- Context window
- 131,072 tokens
- Maximum output
- 131,072 tokens
- Open weights
- Yes
Capability & preference
Human preference · Arena
Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.
Recorded results
Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.
Arena text preference / chinesehuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,249.28 rating
Confidence interval: 1,217.29 – 1,281.28
344 votes
- Category
- chinese
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,248.72 rating
Confidence interval: 1,231.69 – 1,265.76
1286 votes
- Category
- coding
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,209.77 rating
Confidence interval: 1,186.66 – 1,232.89
707 votes
- Category
- creative_writing
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,266.52 rating
Confidence interval: 1,254.19 – 1,278.84
2482 votes
- Category
- english
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,154.19 rating
Confidence interval: 1,141.40 – 1,166.99
4162 votes
- Category
- exclude_ties
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,250.69 rating
Confidence interval: 1,216.56 – 1,284.82
316 votes
- Category
- expert
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,240.37 rating
Confidence interval: 1,229.08 – 1,251.66
3003 votes
- Category
- hard_prompts
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,266.44 rating
Confidence interval: 1,249.93 – 1,282.95
1378 votes
- Category
- hard_prompts_english
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,212.26 rating
Confidence interval: 1,192.82 – 1,231.70
1052 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,212.54 rating
Confidence interval: 1,193.19 – 1,231.89
992 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,225.80 rating
Confidence interval: 1,190.31 – 1,261.28
311 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,252.08 rating
Confidence interval: 1,230.76 – 1,273.40
885 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,275.93 rating
Confidence interval: 1,241.17 – 1,310.69
288 votes
- Category
- industry_mathematical
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,218.63 rating
Confidence interval: 1,177.92 – 1,259.33
259 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,252.77 rating
Confidence interval: 1,239.36 – 1,266.18
2138 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,226.72 rating
Confidence interval: 1,209.04 – 1,244.39
1204 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,221.91 rating
Confidence interval: 1,206.74 – 1,237.07
1607 votes
- Category
- instruction_following
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,241.68 rating
Confidence interval: 1,224.75 – 1,258.60
1311 votes
- Category
- longer_query
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,252.24 rating
Confidence interval: 1,220.52 – 1,283.97
344 votes
- Category
- math
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,242.68 rating
Confidence interval: 1,222.49 – 1,262.88
945 votes
- Category
- multi_turn
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,216.10 rating
Confidence interval: 1,204.92 – 1,227.27
3148 votes
- Category
- non_english
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,241.29 rating
Confidence interval: 1,232.85 – 1,249.72
5630 votes
- Category
- overall
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,168.05 rating
Confidence interval: 1,122.81 – 1,213.30
230 votes
- Category
- polish
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,200.29 rating
Confidence interval: 1,171.54 – 1,229.05
425 votes
- Category
- russian
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating
Granite-4.0-H-Small
1,258.33 rating
Confidence interval: 1,212.36 – 1,304.31
192 votes
- Category
- spanish
- Evaluated configuration
- Granite-4.0-H-Small
- Evaluation type
- human preference
Run it locally
API cost
Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.
Granite-4.0-H-Small
Capabilities
- Context window
- 131,072 tokens
- Maximum output
- 131,072 tokens
- Reasoning
- No
- Tool calling
- Yes
- Open weights
- Yes
- Checkpoint tensor payload
- 59.99 GiB
Input modalities
- text
Output modalities
- text
Processing & platform
- Architecture
- granitemoehybrid
0 selected Compare 0