Navigate
Advertise here

Granite 4.2 8B vs gpt-oss-120b

Granite 4.2 8B leads on average score with 4.4 vs 4.2. gpt-oss-120b has the lower benchmark cost at $0.016 vs $0.037. gpt-oss-120b is faster at 28.63s vs 84.66s, with pass rates of 15.9% vs 34.8%.

Last updated at: 2026-10-07

Compared models

Rank
#355
Total Output Tokens
133,268
Response Time (avg)
84.66s
Total Cost
$0.037
Rank
#361
Total Output Tokens
62,213
Response Time (avg)
28.63s
Total Cost
$0.016
Recommended model gpt-oss-120b

Its score stays close to the best score here (4.2 vs 4.4), while costing about 2.3x less than Granite 4.2 8B.

Detailed comparison

Metric Granite 4.2 8B Granite 4.2 8B none Release: 2026-09-02 gpt-oss-120b gpt-oss-120b none Release: 2025-08-05
Score 4.4 4.2
Rank #355 #361
Reliability 9.1 10.0
Consistency 9.3 7.6
Attempts 69/69 60/69
Tests Correct
Attempt pass rate 15.9% 34.8%
Flaky tests 2 3
Total Runs 69 60
Cost per result 1.206 0.274
Total Cost $0.037 $0.016
Input Price $0.060 / 1M $0.037 / 1M
Output Price $0.250 / 1M $0.170 / 1M
Cache Read Price $0.015 / 1M N/A
Cache Write Price N/A N/A
Total Input Tokens 226,872 131,652
Output Tokens 133,268 16,869
Reasoning Tokens 0 53,081
Response Time (avg) 84.66s 28.63s
Response Time (max) 747.17s 140.90s
Response Time (total) 1947.17s 486.69s
Parameters 8B 117B total (5.1B 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.

#355 IBM: Granite 4.2 8B

none
Provider returned error
Cost
$0.000
Time
0.2s
Tokens
0 tok

#361 gpt-oss-120b

none
No showcase result has been generated for this model yet.
Cost
N/A
Time
-
Tokens
0 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% 0 105.60s 8,358 47,944 0
gpt-oss-120b 1.5 0.4 22.2% 1 9.57s 901 204 3,028

Quick Compare

Switch Comparison Pair

Trinity Large PreviewdefaultFree AvailablevsGranite 4.2 8BnoneGranite 4.2 8BnonevsKAT Coder AIR V2.5defaultCobuddymediumFree AvailablevsGranite 4.2 8Bnonegpt-oss-120bnonevsGLM 4.7 FlashmediumGranite 4.2 8BnonevsRing 2.6 1tdefaultGranite 4.2 8BnonevsLing 2.6 FlashdefaultMiniMax M2.5mediumvsgpt-oss-120bnoneGranite 4.2 8BnonevsMiniMax M2.5mediumgpt-oss-120bnonevsQwen3.5-9BmediumGranite 4.2 8BnonevsNemotron 3.5 LightninglowFree AvailableGranite 4.2 8BnonevsLaguna S 2.1lowFree AvailableGranite 4.2 8BnonevsGLM 4.7 Flashmedium