The LLM benchmark aggregator.

Granite 3.2 8B Instruct

IBM · 8B

Strongest in Toxicity avoidance (#50 of 266), weakest in Safety (#252 of 327). Above par in 2 of 8 scopes. Among the models it meets almost everywhere, it finishes behind Claude Sonnet 4.5 and Claude Haiku 4.5 and ahead of DeepSeek V4 Flash and GLM-5.2.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Agents46.7−19.6#187/2981/6

Tool use42−32#60/811/1

Safety46.8−14.6#252/3271/3

Toxicity avoidance56.6−2.4#50/2661/1

Jailbreak resistance55.9−9.1#102/2661/1

Harm refusal48.7−12.8#172/2901/2

Fairness42.9−27.1#235/2901/2

Secure code26.5−39.7#251/2681/1

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for Granite 3.2 8B Instruct, left for the other.

Nous Hermes 2 Mixtral 8x7B DPOwon 5 · lost 0

SmolLM 1.7B Instructwon 6 · lost 0

H2o Danube3 500M Chatwon 6 · lost 0

SmolLM2 1.7B Instructwon 6 · lost 0

DeepSeek R1 Distill Qwen 7Bwon 6 · lost 0

OLMoE 1B-7B Instructwon 5 · lost 1

Mistral Instruct v0.1 (7B)won 5 · lost 1

OLMo 7B Instructwon 5 · lost 1

DBRX Instructwon 5 · lost 1

LFM2 350Mwon 5 · lost 1

SmolLM2 360M Instructwon 5 · lost 1

CodeLlama 7B Instructwon 5 · lost 1

Qwen2.5 0.5B Instructwon 5 · lost 1

Gemma 2B ITwon 5 · lost 1

GPT-3.5 Turbowon 5 · lost 1

DeepSeek R1 Distill Llama 8Bwon 5 · lost 1

H2o Danube3 4B Chatwon 5 · lost 1

Llama 3.1 70B Instructwon 5 · lost 1

Command R+won 5 · lost 1

SmolLM 360M Instructwon 5 · lost 1

Qwen3 8Bwon 6 · lost 2

DeepSeek R1won 5 · lost 2

Gemma 4 E2Bwon 5 · lost 2

Gemma 4 E4Bwon 5 · lost 2

Magistral Small 1.2won 5 · lost 2

Ling 3.0 Flash VLwon 5 · lost 2

MiMo V2.6 Flashwon 5 · lost 2

RakutenAI 7B Chatwon 4 · lost 2

PowerMoE 3Bwon 4 · lost 2

Phi 2won 4 · lost 2

Gemma 7B ITwon 4 · lost 2

PowerLM 3Bwon 4 · lost 2

Laguna Xs 2.1won 4 · lost 2

RakutenAI 7B Instructwon 4 · lost 2

Falcon Mamba 7B Instructwon 4 · lost 2

GLM-4 9B Chatwon 4 · lost 2

Qwen2 7B Instructwon 4 · lost 2

Magistral Smallwon 4 · lost 2

Jamba Mini 1.7won 4 · lost 2

OLMo 2 7B Instructwon 4 · lost 2

Mixtral 8x7B Instructwon 4 · lost 2

Grok 3won 4 · lost 2

Aya 23 8Bwon 4 · lost 2

Laguna S 2.1won 4 · lost 2

Falcon H1 0.5B Instructwon 4 · lost 2

Zephyr 7B Betawon 4 · lost 2

Magistral Mediumwon 4 · lost 2

Jamba 1.6 Miniwon 4 · lost 2

LFM2 700Mwon 4 · lost 2

Jamba Large 1.7won 4 · lost 2

Jamba 1.5 Largewon 4 · lost 2

DeepSeek LLM 67B Chatwon 4 · lost 2

Qwen2.5 3B Instructwon 4 · lost 2

Granite 3.0 1B A400M Instructwon 4 · lost 2

QwQ 32Bwon 4 · lost 2

Jamba Instructwon 4 · lost 2

Qwen2.5 1.5B Instructwon 4 · lost 2

Gemma 3 4B ITwon 5 · lost 3

Command R7Bwon 5 · lost 3

Qwen3 30B A3B Instructwon 5 · lost 3

Command Awon 5 · lost 3

Gemma 4 31Bwon 4 · lost 3

DeepSeek V3won 4 · lost 3

Mercury 2won 4 · lost 3

Gemini 2.5 Prowon 4 · lost 3

Gemma 4 26B A4Bwon 4 · lost 3

Llama 3.2 3B Instructwon 4 · lost 3

Gemma 4 12Bwon 4 · lost 3

Gemini 3.5 Flashwon 4 · lost 3

Ling 3.0 Flashwon 4 · lost 3

Grok Build 0.1won 4 · lost 3

Gemini 3.6 Flashwon 4 · lost 3

GLM-5.2won 4 · lost 3

DeepSeek V4 Flashwon 4 · lost 3

Mistral Instruct v0.3 (7B)won 3 · lost 3

SEA LION V1 7B ITwon 3 · lost 3

Qwen1.5 14B Chatwon 3 · lost 3

NexusRaven V2 13Bwon 3 · lost 3

Gemma 2 9B ITwon 3 · lost 3

Mistral 7B Instruct v0.2won 3 · lost 3

Mimo V2.6 Pro Ultraspeedwon 3 · lost 3

Jamba 1.5 Miniwon 3 · lost 3

Open Codestral Mambawon 3 · lost 3

LFM2 1.2Bwon 3 · lost 3

Jamba 1.6 Largewon 3 · lost 3

Mixtral 8x22B Instructwon 3 · lost 3

GLM-4.5won 3 · lost 3

Sarvam 1won 3 · lost 3

Qwen2.5 7B Instruct 1Mwon 3 · lost 3

Komodo 7B Basewon 3 · lost 3

Mistral Small 3 24B Instructwon 3 · lost 3

Hunyuan A13B Instructwon 3 · lost 3

Fuguwon 3 · lost 3

DeepSeek V3.1won 3 · lost 3

Granite 3.0 3B A800M Instructwon 3 · lost 3

Granite 3.0 2B Instructwon 3 · lost 3

Qwen2 VL 72Bwon 3 · lost 3

Qwen 2.5 Instruct (7B)won 3 · lost 3

Sarvam 30Bwon 3 · lost 3

Qwen2 57B A14B Instructwon 3 · lost 3

Kimi K2 Thinkingwon 3 · lost 3

Grok 4won 4 · lost 4

Grok 4.1 Fastwon 4 · lost 4

Phi-4 Mini Instructwon 3 · lost 4

Llama 3.3 70B Instructwon 3 · lost 4

GPT OSS 20Bwon 3 · lost 4

Amazon Nova Prowon 3 · lost 4

Qwen3.5 122B A10Bwon 3 · lost 4

NVIDIA Nemotron 3 Super 120B A12Bwon 3 · lost 4

Grok 4.3won 3 · lost 4

Mistral Large 4won 3 · lost 4

Gemini 3.8 Flashwon 3 · lost 4

NVIDIA Nemotron 3 Ultra 550B A55Bwon 3 · lost 4

Kimi K2.7 Codewon 3 · lost 4

MiMo V2.5 Prowon 3 · lost 4

Qwen3.7 Maxwon 3 · lost 4

GLM-5.3 Flashwon 3 · lost 4

Kimi K2.6won 3 · lost 4

MiniMax M3won 3 · lost 4

SeaLLMs V3 7B Chatwon 2 · lost 3

Gemma 3 12B ITwon 3 · lost 5

Llama 3.2 1B Instructwon 3 · lost 5

GPT-4.1 nanowon 3 · lost 5

Amazon Nova Microwon 3 · lost 5

DeepSeek V3.2won 3 · lost 5

Gemini 2.5 Flashwon 3 · lost 5

GLM-4.6won 3 · lost 5

Aya 23 35Bwon 2 · lost 4

Yi 1.5 9B Chatwon 2 · lost 4

Gemma 2 2B ITwon 2 · lost 4

Gemini 2.0 Flash-Litewon 2 · lost 4

Starling LM 7B Betawon 2 · lost 4

Dolphin 2.5 Mixtral 8x7bwon 2 · lost 4

Gemma2 9B Cpt Sea Lionv3 Instructwon 2 · lost 4

Apertus 8B Instructwon 2 · lost 4

Qwen2.5 32B Instructwon 2 · lost 4

LongWriter Glm4 9Bwon 2 · lost 4

Qwen2.5 14B Instruct 1Mwon 2 · lost 4

Gemini 1.5 Flashwon 2 · lost 4

GLM-5 Turbowon 2 · lost 4

Qwen3 VL 235B A22B Instructwon 2 · lost 4

Phi 3 Small 128k Instructwon 2 · lost 4

GPT-5.6 Astrawon 2 · lost 4

Llama 3 8B Instructwon 2 · lost 4

GPT OSS Safeguard 20Bwon 2 · lost 4

EAI Llama 3.1 8B Instruct SageAlignwon 2 · lost 4

Phi-3 Small 8K Instructwon 2 · lost 4

Llama 3.1 Tulu 3 8B SFTwon 2 · lost 4

Llama 3.1 Tulu 3 8Bwon 2 · lost 4

EAI SageAlign 8B LoRAwon 2 · lost 4

Qwen3.5 27Bwon 2 · lost 4

Phi 3.5 MoE Instructwon 2 · lost 4

Titan Text Express V1won 2 · lost 4

Nova Premierwon 2 · lost 4

Fugu Ultra V2won 2 · lost 4

O3 Miniwon 2 · lost 5

Inklingwon 2 · lost 5

GPT-4o miniwon 2 · lost 5

GPT-4won 2 · lost 5

Qwen3.5 397B A17Bwon 2 · lost 5

GPT OSS 120Bwon 2 · lost 5

GPT-6 Lunawon 2 · lost 5

Grok 4.5won 2 · lost 5

GLM-5.1won 2 · lost 5

Qwen3.8 Maxwon 2 · lost 5

Kimi K2.5won 2 · lost 5

GPT-5.6 Terrawon 2 · lost 5

GPT-5.5won 2 · lost 5

DeepSeek V4 Prowon 2 · lost 5

GPT-5.6 Lunawon 2 · lost 5

GPT-5.6 Solwon 2 · lost 5

GPT-6 Solwon 2 · lost 5

Claude Opus 5won 2 · lost 5

Llama 3.1 8B Instructwon 2 · lost 6

Gemma 3 27B ITwon 2 · lost 6

Llama 4 Scout Instructwon 2 · lost 6

Llama 4 Maverick Instructwon 2 · lost 6

GPT-4.1 miniwon 2 · lost 6

Qwen3 235B A22B Instructwon 2 · lost 6

Kimi K2won 2 · lost 6

Sarvam 105Bwon 1 · lost 4

WizardLM 2.8x22Bwon 1 · lost 5

Granite 3.1 2B Instructwon 1 · lost 5

IBM Granite 4.0 Microwon 1 · lost 5

Granite 3.2 2B Instructwon 1 · lost 5

Gemini 1.5 Prowon 1 · lost 5

Llama 3.1 405B Instructwon 1 · lost 5

Reflection Llama 3.1 70Bwon 1 · lost 5

Granite 3.1 3B A800M Instructwon 1 · lost 5

Phi-3 Medium 4K Instructwon 1 · lost 5

Claude 3 Haikuwon 1 · lost 5

Qwen2.5 14B Instructwon 1 · lost 5

Gemini 2.0 Flashwon 1 · lost 5

Phi 3 Medium 128k Instructwon 1 · lost 5

Granite 3.0 8B Instructwon 1 · lost 5

Apertus 70B Instructwon 1 · lost 5

Granite 4.0 H Microwon 1 · lost 5

Granite 4.0 H Tinywon 1 · lost 5

Phi-3 Mini 128K Instructwon 1 · lost 5

Phi 3.5 Mini Instructwon 1 · lost 5

K2 Chatwon 1 · lost 5

Phi-3 Mini 4K Instructwon 1 · lost 5

Llama 3.1 Tulu 3 8B DPOwon 1 · lost 5

Krutrim 2 Instructwon 1 · lost 5

Qwen Maxwon 1 · lost 5

Amazon Nova Litewon 1 · lost 5

Internlm2 Chat 20Bwon 1 · lost 5

Llama 3 8B Instruct RRwon 1 · lost 5

Granite 4.0 H Smallwon 1 · lost 5

Llama 3 8B Instruct MopeyMulewon 1 · lost 5

Llama 3.2 Vision (90B)won 1 · lost 5

Cogito 671B V2 P1won 1 · lost 5

Qwen2.5 72B Instructwon 1 · lost 5

O1 Miniwon 1 · lost 5

GPT-4 Turbowon 1 · lost 5

O1won 1 · lost 5

GPT-4owon 1 · lost 5

Qwen3.5 35B A3Bwon 1 · lost 5

Claude 3.5 Haikuwon 1 · lost 6

MiniMax M2.7won 1 · lost 6

GPT-5.1won 1 · lost 6

GPT-5won 1 · lost 6

Hy3won 1 · lost 6

GPT-5.4won 1 · lost 6

Grok 4.6won 1 · lost 6

Muse Spark 1.1won 1 · lost 6

MiMo V2.6 Prowon 1 · lost 6

Claude Opus 5.5won 1 · lost 6

GPT-5 miniwon 1 · lost 7

O3won 1 · lost 7

GPT-5 nanowon 1 · lost 7

O4 Miniwon 1 · lost 7

GPT-4.1won 1 · lost 7

GPT-5.2won 1 · lost 7

Aya Expanse 32Bwon 0 · lost 5

Llama 2 13B Chatwon 0 · lost 5

Meta SecAlign 8Bwon 0 · lost 5

Palmyra Xwon 0 · lost 5

Gemini 2.5 Flash-Litewon 0 · lost 5

Claude Sonnet 4.6won 0 · lost 5

Gemini 3 Prowon 0 · lost 5

Smaug Llama 3 70B Instructwon 0 · lost 6

SeaLLM 7B V2won 0 · lost 6

Llama 3 70B Instructwon 0 · lost 6

Aya Expanse 8Bwon 0 · lost 6

Llama 3.2 Vision (11B)won 0 · lost 6

Llama 3.1 Nemotron 70B Instructwon 0 · lost 6

Llama 2 70B Chatwon 0 · lost 6

Llama 2 7B Chatwon 0 · lost 6

Qwen2 72B Instructwon 0 · lost 6

Claude 3 Sonnetwon 0 · lost 6

Claude Opus 4won 0 · lost 6

Foundation Sec 8B Instructwon 0 · lost 6

Gemma 2 27B ITwon 0 · lost 6

Apriel 1.5 15B Thinkerwon 0 · lost 6

Mistral NeMo Minitron 8B Instructwon 0 · lost 6

MiniMax M2won 0 · lost 6

Claude 3 Opuswon 0 · lost 6

Claude 3.5 Sonnetwon 0 · lost 6

Claude 3.7 Sonnetwon 0 · lost 7

Claude Sonnet 4won 0 · lost 7

Claude Opus 4.7won 0 · lost 7

Claude Opus 4.8won 0 · lost 7

Claude Sonnet 5won 0 · lost 7

Grok 4.7won 0 · lost 7

Granite 3.1 8B Instructwon 0 · lost 8

Phi 4won 0 · lost 8

Claude Haiku 4.5won 0 · lost 8

Claude Sonnet 4.5won 0 · lost 8

§ 3 · Sources

Where the numbers come from

2 publications, 7 figures. Every one links to the page it was read from.

BFCL ↗1 measureread 2026-11-11

BFCL v4 26.87%

Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-11-11

Enkrypt · Jailbreak risk 7.9%Enkrypt · Harmful content risk 53.9%Enkrypt · CBRN risk 7.3%Enkrypt · Toxicity risk 1.6%Enkrypt · Bias risk 88.9%Enkrypt · Insecure code risk 81.3%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=granite-3-2-8b-instruct)](https://publicai.io/model-index/m/granite-3-2-8b-instruct)