GeekRadar / The comparison desk
COMPARE.
AI models
ibm

ibm-granite-h-small

Model evaluation

Capability & preference

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

ibm-granite-h-small 1241.4 ibm-granite-h-small · 95% interval 1233.0–1249.8 · 5,630 votes · 2026-10-15

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 · 2026-10-15 · rating

ibm-granite-h-small

1,250.07 rating

Confidence interval: 1,218.06 – 1,282.09

343 votes

Category
chinese
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,248.98 rating

Confidence interval: 1,231.97 – 1,265.99

1290 votes

Category
coding
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,210.95 rating

Confidence interval: 1,187.82 – 1,234.08

704 votes

Category
creative_writing
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,266.02 rating

Confidence interval: 1,253.70 – 1,278.34

2485 votes

Category
english
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,154.27 rating

Confidence interval: 1,141.47 – 1,167.07

4161 votes

Category
exclude_ties
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,250.70 rating

Confidence interval: 1,216.56 – 1,284.84

316 votes

Category
expert
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,240.49 rating

Confidence interval: 1,229.20 – 1,251.78

3003 votes

Category
hard_prompts
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,265.66 rating

Confidence interval: 1,249.16 – 1,282.17

1380 votes

Category
hard_prompts_english
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,212.74 rating

Confidence interval: 1,193.32 – 1,232.17

1052 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,213.03 rating

Confidence interval: 1,193.68 – 1,232.39

990 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,224.83 rating

Confidence interval: 1,189.35 – 1,260.31

312 votes

Category
industry_legal_and_government
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,251.86 rating

Confidence interval: 1,230.54 – 1,273.18

886 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,274.96 rating

Confidence interval: 1,240.19 – 1,309.72

289 votes

Category
industry_mathematical
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,218.69 rating

Confidence interval: 1,177.97 – 1,259.41

259 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,252.93 rating

Confidence interval: 1,239.53 – 1,266.34

2142 votes

Category
industry_software_and_it_services
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,227.75 rating

Confidence interval: 1,210.07 – 1,245.44

1199 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,222.34 rating

Confidence interval: 1,207.18 – 1,237.51

1607 votes

Category
instruction_following
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,241.91 rating

Confidence interval: 1,224.98 – 1,258.84

1310 votes

Category
longer_query
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,249.40 rating

Confidence interval: 1,217.64 – 1,281.15

345 votes

Category
math
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,242.93 rating

Confidence interval: 1,222.73 – 1,263.14

944 votes

Category
multi_turn
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,216.65 rating

Confidence interval: 1,205.47 – 1,227.82

3145 votes

Category
non_english
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,241.40 rating

Confidence interval: 1,232.97 – 1,249.84

5630 votes

Category
overall
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,168.09 rating

Confidence interval: 1,122.85 – 1,213.34

230 votes

Category
polish
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,199.94 rating

Confidence interval: 1,171.19 – 1,228.70

426 votes

Category
russian
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

ibm-granite-h-small

1,258.22 rating

Confidence interval: 1,212.24 – 1,304.21

192 votes

Category
spanish
Evaluated configuration
ibm-granite-h-small
Evaluation type
human preference

Run it locally

Inference · memory estimate

ibm-granite-h-small

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Apache 2.0