Skip to content

IBM · Chat

Granite 4.2 8B

IBM 🇺🇸ChatOpen weights. One of its sources says the weights are publicly available.
From Aiml API's listing:
Granite 4.2 8B is a dense reasoning model from IBM, fine-tuned from Granite 4.1 8B for mathematics, code generation, multilingual dialogue and agentic workflows that need multi-step reasoning. Reasoning is on by default and its effort is configurable (high, low or none), and the model supports tool calling and structured outputs over a 131K-token context.
llminfo IQ
110.9
Market median price
$0.1075 blended / 1M
Max context
131K
Listings
22

Evidence and sources

🤗 ibm-granite/granite-4.2-8bllminfo IQ 110.9, ranked among chat models. From 57 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 90.2, 95% interval 73–108. From 8 benchmark families across 2 sources. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedMaker HQ: US (source)
All rankings: #206 on OpenRouter today #240 on OpenRouter this week #247 on OpenRouter this month #370 on NanoGPT recentlySources: LMArena data, Models.dev data, AI/ML API dataPrimary data from LMArena from Models.dev: weights availability from AI/ML API: primary mode, description

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Advantages

Open weights available

The weights are public. 4 of the 8 peers that say either way have open weights.

Trade-offs

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 6 maker's prices; 1 official partner's price; 1 market median price.

Input & output

as stated by the model's providers
Text
Text

Capabilities

Supported operations and provider-stated features

ReasoningFunction callingStructured outputPrompt caching

Listings

 Eden AIInference by DeepInfradeepinfra/ibm-granite/granite-4.2-8bSource: Models.dev data131K$0.06cache read $0.015$0.25——
 Fastrouteribm/granite-4.2-8bSource: FastRouter data131K$0.06cache read $0.015$0.25——
 Flowdotibm-granite/granite-4.2-8bSource: FlowDot data131K$0.06$0.25——
 Hugging FaceInference by Coreweaveibm-granite/granite-4.2-8bSource: Hugging Face Inference Providers data131K$0.06$0.15——
 Kiloibm-granite/granite-4.2-8bSources: Kilo AI Gateway data, Models.dev data131K$0.06cache read $0.015$0.15——
 LlmgatewayInference by DeepInfraibm-granite/granite-4.2-8bSource: LLMGateway dataNot published by the provider$0.06cache read $0.015$0.25——
 NanoGPTibm-granite/granite-4.2-8bSources: Models.dev data, NanoGPT131K$0.06cache read $0.015$0.15——
 Nous Portalibm-granite/granite-4.2-8bSource: Nous Portal Inference Gateway data131K$0.06cache read $0.015$0.15——
 OpenRouterInference by Coreweaveibm-granite/granite-4.2-8b · bf16Source: OpenRouter data docs131K$0.06cache read $0.015$0.1569270ms
 Replicateibm-granite/granite-4.2-8bSources: LiteLLM data, Replicate data docsNot published by the provider$0.06$0.25——
Show all 19 listings

Showing 10 of 19: the cheapest per 1M input tokens from each provider.

Also sold by Llmgateway Providers, Weights & Biases.

3 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Human preference

Arena ELO31 related benchmarks · 31 results
Arena ELO
Source: LMArena data · Nov 2026
Arena-Hard v2
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Chinese
Source: LMArena data · Nov 2026
Creative Writing
Source: LMArena data · Nov 2026
English
Source: LMArena data · Nov 2026
Exclude Ties
Source: LMArena data · Nov 2026
Show 25 more resultsHide 25 results
Expert
Source: LMArena data · Nov 2026
Hard Prompts
Source: LMArena data · Nov 2026
Hard Prompts English
Source: LMArena data · Nov 2026
Longer Query
Source: LMArena data · Nov 2026
Multi Turn
Source: LMArena data · Nov 2026
Non English
Source: LMArena data · Nov 2026

Coding

SciCodeSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Agentic

BFCL-V4Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Instruction following

IFBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Long context

Math

AIME 2025Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
HMMT25Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Reasoning

ProfBenchSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Knowledge

MMLU-ProSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Science

GPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Other

Data credits: Data by LLM Stats · LMArena, CC BY 4.0 (licence) · Baromètre de l’écriture IA, Johannes Eva; public numeric ranking data, © 2026 Johannes Eva, no results licence stated

Also known as: deepinfra/granite-4.2-8b, deepinfra/ibm-granite/granite-4.2-8b, ibm-granite/granite-4.2-8b, ibm/granite-4.2-8b, replicate/ibm-granite/granite-4.2-8b, wandb/ibm-granite/granite-4.2-8b

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works