IBMOpen weightsReleased 31 Aug 2026

Granite 4.2 8B benchmarks, speed and API pricing

Reviewed by LLM Verifications team · Tested 26 Oct 2026

Compare

Available through OpenRouter on

CoreWeave$0.06 in · $0.15 out
DeepInfra$0.06 in · $0.25 out

Intelligence36.2% correct#60 of 72

Speed60tok/s#53 of 72

Cost$0.000077per question#3 of 72List $0.06 in · $0.25 out per 1M

Verbosity227tokens per question#5 of 72

Summary

Answers 29 of 80 verification questions correctly (36.2%), placing #60 of 72 models tested on the current suite.
Generates 60 output tokens per second, faster than 27% of the models we measure.
Starts responding after 0.6 s (median time to first token), including thinking time.
Costs $0.000077 per question as billed, #3 cheapest of 12 models priced under $0.15 per 1M tokens.

Specifications

Reasoning: Yes
Input: text
Output: text
Context window: 131k tokens
Weights: Open weights
Knowledge cutoff: —
OpenRouter ID: ibm-granite/granite-4.2-8b
Last tested: 26 Oct 2026

Peer group: 13 models with a blended list price under $0.15 per 1M tokens (3 input : 1 output). Charts below compare Granite 4.2 8B with all models; cost and speed charts focus on this peer group.

Intelligence

Verification score: the share of 80 questions from open datasets answered correctly. See the benchmark datasets.

Score vs. cost per question

Granite 4.2 8B circled; dashed line joins models no cheaper model beats

Score vs. end-to-end response time

Granite 4.2 8B circled · accurate and quick sits top left

Score vs. tokens per question

Granite 4.2 8B circled · output tokens, reasoning included

Verification score

All models tested on the current suite

61 more models · see the full leaderboard

Cost

Average cost of one question, split into input, reasoning and answer tokens at each model's list price. Reasoning tokens are billed as output.

Cost per question, by token type

Peer group: blended price under $0.15

1gpt-oss-20b$0.000049
2Granite 4.0 Micro$0.000051
3Granite 4.2 8B$0.000077
4Nova Micro 1.0$0.000087
5Mercury 2.5$0.00011
6Phi 4$0.00012
7Ministral 3 3B 2512$0.0002
8Solar Mini 4$0.00053
9Nemotron 3.5 Lightning$0.00059
10Ling 3.1 Flash$0.00061
11Reka Flash 3$0.0008
12Gemma 4 26B A4B$0.0011

Input Reasoning Answer

Cost per question, as billed

All models · lower is better

1Granite 4.0 Micro$0.000051
2gpt-oss-20b$0.000066
3Granite 4.2 8B$0.000077
4Nova Micro 1.0$0.000087
5Mercury 2.5$0.00011
6Phi 4$0.00012
7gpt-oss-120b$0.00013
8GLM 5.3 Flash$0.00015
9Ministral 3 3B 2512$0.00019
10GPT-6 Luna$0.00025

62 more models · see the full leaderboard

Speed and latency

Median of 3 streamed runs of the same prompt, measured 25 Oct 2026.

Output speed

Peer group (under $0.15) · higher is better

Time to first token

Peer group (under $0.15) · lower is better

End-to-end response time

Peer group (under $0.15) · lower is better

Latency vs. output speed

Granite 4.2 8B circled · quick to start and fast sits top left

Results by category

Questions answered correctly in each category

Science13/30

Engineering4/15

Medicine1/11

Economics5/9

Law2/6

Humanities & social science4/9

More from IBM

IBM#69Granite 4.0 Micro25.0% verification score38 tok/s0.7 s first token$0.000051 / question

All IBM models →

Compare Granite 4.2 8B

Head to head with its closest rivals, other IBM models and the top models overall.

Hosts and prices

Every host serving this model through OpenRouter, from OpenRouter's catalogue

Host	Input / 1M	Output / 1M	Context	Max output	Quantization
CoreWeave	$0.06	$0.15	131k	118k	bf16
DeepInfra	$0.06	$0.25	131k	118k	bf16

Sources

Scores, speed and billed cost are our own measurements. Prices, context window and hosts come from OpenRouter's catalogue on the test date.

Granite 4.2 8B on OpenRouter: list prices and hosts
Model card on Hugging Face: weights and licence
SuperGPQA (paper): 50 of the questions, ODC-BY 1.0
MMLU-Pro (paper): 30 of the questions, MIT
OpenRouter generation records: where billed cost and native token counts come from
Our methodology and editorial standards

Granite 4.2 8B: frequently asked questions

How accurate is Granite 4.2 8B?: Granite 4.2 8B answered 29 of 80 questions correctly (36.2%) in our 2026-10-27 run, ranking #60 of 78 models.
How fast is Granite 4.2 8B?: Granite 4.2 8B generated a median of 60 output tokens per second in our 2026-10-27 run, ranking #53 of 78. Median time to first token was 0.6 s.
What is the latency of Granite 4.2 8B?: Granite 4.2 8B had a median time to first token of 0.6 s in our 2026-10-27 run, ranking #25 of 78 models.
How much does Granite 4.2 8B cost?: Granite 4.2 8B cost $0.000077 per question on average as billed in our 2026-10-27 run, reasoning tokens included.
What is Granite 4.2 8B API pricing?: Granite 4.2 8B is listed at $0.06 per million input tokens and $0.25 per million output tokens on OpenRouter.
What is the context window of Granite 4.2 8B?: Granite 4.2 8B accepts up to 131,072 tokens of context, according to OpenRouter's model catalogue.
Is Granite 4.2 8B open source?: Yes. Granite 4.2 8B is an open-weights model: IBM publishes the weights, so you can run it yourself as well as through an API.
Who makes Granite 4.2 8B?: Granite 4.2 8B is made by IBM. We tested it through OpenRouter using the model ID ibm-granite/granite-4.2-8b.
Where can I use Granite 4.2 8B via API?: Through OpenRouter, Granite 4.2 8B is served by 1 host: DeepInfra.

Summary

Answers 29 of 80 verification questions correctly (36.2%), placing #60 of 72 models tested on the current suite.

Generates 60 output tokens per second, faster than 27% of the models we measure.

Starts responding after 0.6 s (median time to first token), including thinking time.

Costs $0.000077 per question as billed, #3 cheapest of 12 models priced under $0.15 per 1M tokens.

Host

Input / 1M

Output / 1M

Context

Max output

Quantization

CoreWeave

$0.06

$0.15

131k

118k

bf16

DeepInfra

$0.06

$0.25

131k

118k

bf16

Sources

Scores, speed and billed cost are our own measurements. Prices, context window and hosts come from OpenRouter's catalogue on the test date.

Granite 4.2 8B: frequently asked questions

How accurate is Granite 4.2 8B?

Granite 4.2 8B answered 29 of 80 questions correctly (36.2%) in our 2026-10-27 run, ranking #60 of 78 models.

How fast is Granite 4.2 8B?

Granite 4.2 8B generated a median of 60 output tokens per second in our 2026-10-27 run, ranking #53 of 78. Median time to first token was 0.6 s.

What is the latency of Granite 4.2 8B?

Granite 4.2 8B had a median time to first token of 0.6 s in our 2026-10-27 run, ranking #25 of 78 models.

How much does Granite 4.2 8B cost?

Granite 4.2 8B cost $0.000077 per question on average as billed in our 2026-10-27 run, reasoning tokens included.

What is Granite 4.2 8B API pricing?

Granite 4.2 8B is listed at $0.06 per million input tokens and $0.25 per million output tokens on OpenRouter.

What is the context window of Granite 4.2 8B?

Granite 4.2 8B accepts up to 131,072 tokens of context, according to OpenRouter's model catalogue.

Is Granite 4.2 8B open source?

Yes. Granite 4.2 8B is an open-weights model: IBM publishes the weights, so you can run it yourself as well as through an API.

Who makes Granite 4.2 8B?

Granite 4.2 8B is made by IBM. We tested it through OpenRouter using the model ID ibm-granite/granite-4.2-8b.

Where can I use Granite 4.2 8B via API?

Through OpenRouter, Granite 4.2 8B is served by 1 host: DeepInfra.