Granite 4.2 released
IBM · Model release
Source confirmed
Data as of October 15, 2026 · How the score is built
CurrentOpen WeightReasoning
Released Aug 25, 2026128K contextIBM Granite 4.2 8B model card
Granite 4.2 8B
Decision readingGranite 4.2 8B scores 33.4 out of 100 and ranks #170 of 224. This profile shows 14 source-displayable benchmark rows; its strongest eligible category is Instruction Following at #41. Published weights can be self-hosted, but infrastructure cost varies and is not a comparable API token rate.
Released Aug 25, 2026 — see all recent releases
Granite 4.2 8B will be repriced, updated or retired. Get each notice with its source and date. Follow model changes
Decision snapshot
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Capability
33.4/100
field median 49.8#170 of 224 ranked models
Public
#170of 224
Verified —
Price
Self-hosted; infrastructure cost varies
input median $0.75No comparable first-party hosted token rate
Speed
123tok/s
field median 92 tok/sFirst token 17 s
Context
128Ktokens
field median 256,000Maximum output length is tracked separately
Strongest published evidence
Instruction Following ranks #41. A well-rounded choice across a range of tasks.
Validate before choosing
14 published rows leave some tracked benchmark slots empty. No comparable first-party API token rate is published.
Source-linked · 14 displayable benchmark rows
Follow model changesCategory score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticRank #113 of 127Percentile 11thWeight 22%3 benchmarksVerified | 15.7 | 3 benchmarks | Verified | |||
| CodingRank #125 of 150Percentile 17thWeight 20%6 benchmarksVerified | 17.2 | 6 benchmarks | Verified | |||
| ReasoningWeight 17%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MultimodalWeight 12%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| KnowledgeRank #154 of 184Percentile 16thWeight 12%2 benchmarksVerified | 30.6 | 2 benchmarks | Verified | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingRank #41 of 146Percentile 72ndWeight 5%1 benchmarkVerified | 86.5 | 1 benchmark | Verified | |||
| MathWeight 5%2 benchmarksVerified | Score pending | 2 benchmarks | Verified |
14 of 680 tracked benchmark slots have displayable evidence · bars run 0–100
Coverage detailsHow much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- Agentic3/3 verified
- Coding6/6 verified
- ReasoningNot measured
- MultimodalNot measured
- Knowledge2/2 verified
- MultilingualNot measured
- Inst. Following1/1 verified
- Math2/2 verified
Verified sourceProvisionalNot measured
Capability shape
Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.
Granite 4.2 8B category percentile values
- Agentic11th percentile
- Coding17th percentile
- ReasoningNot eligible
- MultimodalNot eligible
- Knowledge16th percentile
- MultilingualNot eligible
- Instruction following72nd percentile
- MathNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- Agentic#113/127
- Coding#125/150
- ReasoningNot ranked
- MultimodalNot ranked
- Knowledge#154/184
- MultilingualNot ranked
- Inst. Following#41/146
- MathNot ranked
Top decileTop quartileMid-fieldNot eligible
Benchmark ledger
Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
Coding6 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SWE-bench Pro | Score19.1% | Versus best verified row Best verified: Claude Opus 5.5 · 89.9% | Gap70.8 behind | Weight26% ref. weight | Provider exact |
| SciCodeScientific Code Benchmark | Score36.1% | Versus best verified row Best verified: Sakana Fugu · 60.1% | Gap24 behind | Weight10% ref. weight | Provider exact |
| SWE Multilingual | Score30.8% | Versus best verified row Best verified: Claude Opus 5.5 · 93.9% | Gap63.1 behind | Weight5% ref. weight | Provider exact |
| Terminal-Bench 2.1 | Score20.6% | Versus best verified row Best verified: SWE-2 · 92.8% | Gap72.2 behind | WeightScored in Agentic | Provider exact |
| SWE-bench VerifiedSoftware Engineering Benchmark Verified | Score47.7% | Versus best verified row Best verified: Claude Opus 5 · 96% | Gap48.3 behind | WeightDisplay only | Provider exact |
| LiveCodeBench v6 | Score73.2% | Versus best verified row Best verified: Sakana Fugu-Ultra · 93.2% | Gap20 behind | WeightDisplay only | Provider exact |
Agentic3 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| Terminal-Bench 2.1 | Score20.6% | Versus best verified row Best verified: SWE-2 · 92.8% | Gap72.2 behind | Weight8% ref. weight | Provider exact |
| BFCL v4Berkeley Function Calling Leaderboard v4 | Score52.4% | Versus best verified row Best verified: BTL-3 · 88.5% | Gap36.1 behind | Weight3% ref. weight | Provider exact |
| τ³-bench resultsτ³-Bench Tool-Agent-User Evaluation | Score58.1% | Versus best verified row Best verified: Mercury 2.5 · 96.0% | Gap37.9 behind | Weight2% ref. weight | Provider exact |
Knowledge2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| MMLU-ProMassive Multitask Language Understanding Professional | Score74.0% | Versus best verified row Best verified: Qwen3.7 Max · 89.6% | Gap15.6 behind | Weight6% ref. weight | Provider exact |
| GPQAGraduate-Level Google-Proof Q&A | Score64.1% | Versus best verified row Best verified: GPT-6 Astra · 96% | Gap31.9 behind | Weight3% ref. weight | Provider exact |
Inst. Following1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| IFBenchInstruction Following Benchmark | Score79.3% | Versus best verified row Best verified: MAI-Thinking-1 · 85% | Gap5.7 behind | WeightWeighted 70% | Provider exact |
Math2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| AIME 2025American Invitational Mathematics Examination 2025 | Score86.7% | Versus best verified row Best verified: GPT-5.2 · 100% | Gap13.3 behind | WeightDisplay only | Provider exact |
| HMMT Feb 2025Harvard-MIT Mathematics Tournament February 2025 | Score78.3% | Versus best verified row Best verified: GLM-4.7 · 97.1% | Gap18.8 behind | WeightDisplay only | Provider exact |
Bars run 0–100; the dark tick marks the best source-verified value
All 14 rowsLineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Radar
Full release historyGranite 4.2 8B release history
Radar confirmed these at the source. Use Granite 4.2 8B in your work? Explore Radar to follow supported changes and choose your alerts.
Radar
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- ibm-granite/granite-4.2-8bIBM Granite 4.2 8B model card
- Context window
- 128KIBM Granite 4.2 8B model card
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not sourced yet
- Availability
- IBM publishes the dense 8B BF16 checkpoint under Apache 2.0 for self-hosted inference. OpenRouter exposes the exact ibm-granite/granite-4.2-8b route with a 131,072-token context window, optional reasoning, structured output, and tool calling. IBM also documents an opt-in long-context extension to 512K; the catalog stores the native and hosted-route 128K limit.
- Cloud regions
- Not tracked yet
- Lifecycle
- Current
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Not documented in the pricing recordIBM Granite 4.2 8B model card
- Self-host
- Open weights available; hardware estimate not sourced
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
Granite 4.2 8B ranks #170 of 224 on the public leaderboard with a score of 33.4/100. It does not yet have enough sourced coverage for a verified position.
Granite 4.2 8B is a open weight model with a 128K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
IBM publishes the dense 8B BF16 checkpoint under Apache 2.0 for self-hosted inference. OpenRouter exposes the exact ibm-granite/granite-4.2-8b route with a 131,072-token context window, optional reasoning, structured output, and tool calling. IBM also documents an opt-in long-context extension to 512K; the catalog stores the native and hosted-route 128K limit.
Granite 4.2 8B sits in the Granite 4.2 family with Granite 4.2 30B, Granite 4.2 3B. 14 of 680 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Instruction Following at #41, while its lowest eligible position is Knowledge at #154. a well-rounded choice across a range of tasks.
Last updated October 15, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Published weights are available, but BenchLM does not yet have a sourced parameter and VRAM profile for this exact model. Hardware cost estimates stay unavailable until that sizing record is complete.
Estimate VRAM from known parametersQuestions
How does Granite 4.2 8B perform overall in AI benchmarks?
Granite 4.2 8B ranks #170 out of 224 models on the public BenchAlign leaderboard, with a score of 33.4/100. Its evidence status is Estimated, and this profile shows 14 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is Granite 4.2 8B good for knowledge and understanding?
Granite 4.2 8B ranks #154 out of 184 eligible models for knowledge and understanding, with a public category score of 30.6/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Granite 4.2 8B good for coding and programming?
Granite 4.2 8B ranks #125 out of 150 eligible models for coding and programming, with a public category score of 17.2/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Granite 4.2 8B good for mathematics?
Granite 4.2 8B has source-displayable benchmark coverage for mathematics, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Granite 4.2 8B good for agentic tool use and computer tasks?
Granite 4.2 8B ranks #113 out of 127 eligible models for agentic tool use and computer tasks, with a public category score of 15.7/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Granite 4.2 8B good for instruction following?
Granite 4.2 8B ranks #41 out of 146 eligible models for instruction following, with a public category score of 86.5/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Granite 4.2 8B open source?
Granite 4.2 8B is an open-weight model from IBM. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.
Which sibling models are related to Granite 4.2 8B?
Granite 4.2 8B belongs to the Granite 4.2 family. Related tracked variants include Granite 4.2 30B, Granite 4.2 3B. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
Does Granite 4.2 8B have full benchmark coverage on BenchLM?
No. Granite 4.2 8B currently has 28 source-displayable rows across 680 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Granite 4.2 8B?
Granite 4.2 8B has a documented context window of 128K. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.