Navigate
Advertise here

Granite 4.2 8B (medium) vs Beam 1.0 501B A23B (high)

The average score is effectively tied at 6.1 vs 6.1. Granite 4.2 8B (medium) has the lower benchmark cost at $0.029 vs $0.674. Beam 1.0 501B A23B (high) is faster at 9.03s vs 50.47s, with pass rates of 37.7% vs 59.4%.

Last updated at: 2026-10-17

Compared models

Rank
#272
Total Output Tokens
50,045
Response Time (avg)
50.47s
Total Cost
$0.029
Rank
#266
Total Output Tokens
147,207
Response Time (avg)
9.03s
Total Cost
$0.674
Recommended model Granite 4.2 8B (medium)

It has the best score here (6.1), while costing about 23.8x less than Beam 1.0 501B A23B (high).

Detailed comparison

Metric Granite 4.2 8B Granite 4.2 8B medium Release: 2026-09-02 Beam 1.0 501B A23B Beam 1.0 501B A23B high Release: 2026-10-17
Score 6.1 6.1
Rank #272 #266
Reliability 8.9 10.0
Consistency 9.7 8.3
Attempts 69/69 69/69
Tests Correct
Attempt pass rate 37.7% 59.4%
Flaky tests 1 5
Total Runs 69 69
Cost per result 0.354 6.126
Total Cost $0.029 $0.674
Input Price $0.060 / 1M $1.000 / 1M
Output Price $0.250 / 1M $3.000 / 1M
Cache Read Price $0.015 / 1M $0.100 / 1M
Cache Write Price N/A N/A
Total Input Tokens 259,708 232,218
Output Tokens 16,567 4,944
Reasoning Tokens 33,580 142,263
Response Time (avg) 50.47s 9.03s
Response Time (max) 557.96s 36.04s
Response Time (total) 1160.91s 207.67s
Parameters 8B 501B total (23B active)
Availability Open source Open source

Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.

Model generation showcase

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#272 IBM: Granite 4.2 8B

medium
Provider returned error
Cost
$0.000
Time
0.2s
Tokens
0 tok

#266 Beam 1.0 501B A23B

high
Cost
$0.008
Time
15.6s
Tokens
6,162 tok

Top Models by Score

Score vs Total Cost

Response Time (avg)

Score vs Response Time (avg)

Total Output Tokens

Score vs Total Output Tokens

Category Breakdown

Coding Score Consistency Attempt pass rate Flaky tests Tests Correct Response Time (avg) Input Tokens Output Tokens Reasoning Tokens
Granite 4.2 8B 5.5 10.0 33.3% 0 12.09s 8,439 2,445 3,900
Beam 1.0 501B A23B 8.4 7.4 88.9% 1 21.05s 7,800 559 44,608

Quick Compare

Switch Comparison Pair

Trinity Large ThinkingmediumvsBeam 1.0 501B A23BhighGranite 4.2 8BmediumvsKAT-Coder-Pro V2.5defaultApodex 1.1 MinilowFree AvailablevsBeam 1.0 501B A23BhighBeam 1.0 501B A23BhighvsInklingnoneFree AvailableQwen3.5 Plus 2026-04-20nonevsBeam 1.0 501B A23BhighGranite 4.2 8BmediumvsQwen3.5 Plus 2026-04-20noneGranite 4.2 8BmediumvsInklingnoneFree AvailableApodex 1.1 MinilowFree AvailablevsGranite 4.2 8BmediumGemini 3.5 FlashnonevsBeam 1.0 501B A23BhighBeam 1.0 501B A23BhighvsStep 5 PreviewmediumKAT-Coder-Pro V2.5defaultvsBeam 1.0 501B A23BhighGemini 3.5 FlashnonevsGranite 4.2 8Bmedium