IBM model product
Granite 3.3 8B
3 8B is an IBM large language model with enhanced reasoning, Fill-in-the-Middle code completion, and a 128K context length.
Updated Oct 23, 2026. Default version: Granite 3.3 8B Instruct
LLMBoard Score1.0Granite 3.3 8B Instruct
Coverage80%14 benchmark families
Context window128KTokens
Official input priceN/AOfficial price unavailable
Granite 3.3 8B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Granite 3.3 8B Benchmark Results
Benchmark scores for Granite 3.3 8B Instruct.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkAlpacaEval 2.0 | Score62.68% | Rank02 | Participants4 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkAttaQ | Score88.50% | Rank02 | Participants3 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkPopQA | Score26.17% | Rank02 | Participants3 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkTruthfulQA | Score66.86% | Rank03 | Participants18 | Percentile88.24% | EvidenceC | Evaluated |
| BenchmarkHumanEval+ | Score86.09% | Rank04 | Participants10 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkBIG-Bench Hard | Score69.13% | Rank14 | Participants21 | Percentile35.00% | EvidenceC | Evaluated |
| BenchmarkHumanEval | Score89.73% | Rank14 | Participants66 | Percentile80.00% | EvidenceC | Evaluated |
| BenchmarkArena Hard | Score57.56% | Rank16 | Participants27 | Percentile42.31% | EvidenceC | Evaluated |
| BenchmarkDROP | Score59.36% | Rank25 | Participants30 | Percentile17.24% | EvidenceC | Evaluated |
| BenchmarkAIME 2024 | Score81.20% | Rank26 | Participants54 | Percentile52.83% | EvidenceC | Evaluated |
| BenchmarkMATH-500 | Score69.02% | Rank33 | Participants33 | Percentile0.00% | EvidenceC | Evaluated |
| BenchmarkGSM8k | Score80.89% | Rank38 | Participants48 | Percentile21.28% | EvidenceC | Evaluated |
| BenchmarkIFEval | Score74.82% | Rank64 | Participants70 | Percentile8.70% | EvidenceC | Evaluated |
| BenchmarkAA AIME 2025 | Score6.67% | Rank92 | Participants96 | Percentile4.21% | EvidenceB | EvaluatedN/A |
| BenchmarkMMLU | Score65.54% | Rank93 | Participants103 | Percentile9.80% | EvidenceC | Evaluated |
| BenchmarkAA CritPt | Score0.00% | Rank139 | Participants200 | Percentile30.65% | EvidenceB | EvaluatedN/A |
| BenchmarkAA IFBench | Score22.45% | Rank164 | Participants168 | Percentile2.40% | EvidenceB | EvaluatedN/A |
| BenchmarkAA HLE Text No Tools | Score4.17% | Rank178 | Participants200 | Percentile11.06% | EvidenceB | EvaluatedN/A |
| BenchmarkAA Omniscience Accuracy | Score9.65% | Rank188 | Participants201 | Percentile6.50% | EvidenceB | EvaluatedN/A |
| BenchmarkAA GPQA Diamond | Score33.84% | Rank193 | Participants199 | Percentile3.03% | EvidenceB | EvaluatedN/A |
| BenchmarkLM Arena Text | Score1,149.55 rating | Rank215 | Participants218 | Percentile1.38% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,208.64 rating | Rank216 | Participants218 | Percentile0.92% | EvidenceA | Evaluated |
Granite 3.3 8B Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext style control | Categorycoding | Rank204 | Rating / score1,288.29 | Votes478 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank205 | Rating / score1,180.49 | Votes320 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank205 | Rating / score1,226.25 | Votes320 | ObservationsN/A | Result date |
| Arenatext style control | Categoryexpert | Rank207 | Rating / score1,236.03 | Votes237 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank209 | Rating / score1,145.09 | Votes211 | ObservationsN/A | Result date |
| Arenatext | Categoryexpert | Rank209 | Rating / score1,141.53 | Votes237 | ObservationsN/A | Result date |
| Arenatext | Categorycoding | Rank210 | Rating / score1,186.20 | Votes478 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank210 | Rating / score1,170.44 | Votes215 | ObservationsN/A | Result date |
| Arenatext | Categorymath | Rank210 | Rating / score1,151.55 | Votes382 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank210 | Rating / score1,219.05 | Votes211 | ObservationsN/A | Result date |
| Arenatext style control | Categoryhard prompts english | Rank210 | Rating / score1,249.85 | Votes517 | ObservationsN/A | Result date |
| Arenatext style control | Categorymath | Rank210 | Rating / score1,190.19 | Votes382 | ObservationsN/A | Result date |
| Arenatext style control | Categoryenglish | Rank211 | Rating / score1,248.53 | Votes1,737 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry legal and government | Rank211 | Rating / score1,249.71 | Votes215 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry software and it services | Rank211 | Rating / score1,250.77 | Votes780 | ObservationsN/A | Result date |
| Arenatext style control | Categorylonger query | Rank211 | Rating / score1,232.89 | Votes461 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry business and management and financial operations | Rank212 | Rating / score1,225.54 | Votes316 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts english | Rank213 | Rating / score1,163.60 | Votes517 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry business and management and financial operations | Rank213 | Rating / score1,142.00 | Votes316 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank213 | Rating / score1,161.97 | Votes780 | ObservationsN/A | Result date |
| Arenatext | Categorylonger query | Rank213 | Rating / score1,161.64 | Votes461 | ObservationsN/A | Result date |
| Arenatext style control | Categoryhard prompts | Rank213 | Rating / score1,231.45 | Votes798 | ObservationsN/A | Result date |
| Arenatext | Categoryenglish | Rank214 | Rating / score1,189.21 | Votes1,737 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts | Rank215 | Rating / score1,145.40 | Votes798 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry entertainment and sports and media | Rank215 | Rating / score1,111.12 | Votes537 | ObservationsN/A | Result date |
| Arenatext | Categoryinstruction following | Rank215 | Rating / score1,130.67 | Votes1,258 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank215 | Rating / score1,149.55 | Votes3,090 | ObservationsN/A | Result date |
| Arenatext style control | Categoryinstruction following | Rank215 | Rating / score1,192.97 | Votes1,258 | ObservationsN/A | Result date |
| Arenatext | Categorycreative writing | Rank216 | Rating / score1,128.67 | Votes450 | ObservationsN/A | Result date |
| Arenatext | Categoryexclude ties | Rank216 | Rating / score992.24 | Votes2,075 | ObservationsN/A | Result date |
Granite 3.3 8B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- N/A
- Tracked offerings
- 0
No provider prices
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Granite 3.3 8B Runtime Performance
Provider-specific output speed and catalog latency for Granite 3.3 8B Instruct. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Granite 3.3 8B Specifications
Technical details for the model's default version.
- Version
- Granite 3.3 8B Instruct
- Released
- Apr 16, 2025
- Knowledge cutoff
- Apr 1, 2024
- Parameters
- 8B
- Context window
- 128K
- Max output
- 8.2K
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- Apache 2.0
Granite 3.3 8B Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGranite 3.3 8B Base | Released | LLMBoardN/A | Parameters8.2B | ContextN/A | Max outputN/A | Open weightsYes | LicenseApache 2.0 |
| VersionGranite 3.3 8B Instruct | Released | LLMBoard1.00 | Parameters8B | Context128K | Max output8.2K | Open weightsYes | LicenseApache 2.0 |
Models similar to Granite 3.3 8B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
IB
IBM Granite 4.0 Tiny
IBM
IB
IBM Granite 4.2 3B
IBM
IB
IBM Granite 4.2 8B
IBM
IB
IBM Granite 4.2 30B
IBM
AC
Qwen3.5 2B
Alibaba Cloud / Qwen Team
GO
Google
Gemini Diffusion
What is Granite 3.3 8B?
Key information about Granite 3.3 8B and its available data.
0 license. It supports retrieval-augmented generation, function calling, response-length and originality controls, and long-context problem-solving.
Data as of 2026-10-22.
FAQ
Common questions about Granite 3.3 8B.
When was Granite 3.3 8B released?
Granite 3.3 8B's default version was released on Apr 16, 2025.
How much does Granite 3.3 8B cost?
No official standard PAYG price is currently available for Granite 3.3 8B.
Who created Granite 3.3 8B?
Granite 3.3 8B was created by IBM.
What is the context window for Granite 3.3 8B?
The default version has a 128K token context window.
Is Granite 3.3 8B open weight?
Yes. The default version is marked as open weight under Apache 2.0.
How many API providers offer Granite 3.3 8B?
No provider offering is currently linked to the default version.