Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

IBM Granite 4.0 Small (with guardian)

IBM

Strongest in Harm refusal (#53 of 299, on 1 of its 2 boards), weakest in Safety (#194 of 336). Above par in 2 of 4 scopes. Among the models it meets almost everywhere, it finishes behind Claude 3.5 Sonnet and Claude 3 Opus and ahead of Llama 2 7B Chat Hf and Granite 4.0 H Small.

§ 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.

Safety49.6#194/336
Harm refusal55.9#53/299
Safe-prompt compliance26#79/82
Fairness51#100/299

§ 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 IBM Granite 4.0 Small (with guardian), left for the other.

OLMo 7B Instruct Hfwon 3 · lost 0
SmolLM 1.7B Instructwon 3 · lost 0
Qwen2.5 7B Instructwon 3 · lost 0
Gemma 7B Itwon 3 · lost 0
H2o Danube3 500M Chatwon 3 · lost 0
Smollm2 1.7B Instructwon 3 · lost 0
Grok 4.1 Fastwon 3 · lost 0
LFM2 350Mwon 3 · lost 0
Qwen3 30B A3B Instructwon 3 · lost 0
Laguna Xs 2.1won 3 · lost 0
RakutenAI 7B Instructwon 3 · lost 0
Falcon Mamba 7B Instructwon 3 · lost 0
GLM-4 9B Chatwon 3 · lost 0
Nous Hermes 2 Mixtral 8x7b Dpowon 3 · lost 0
CodeLlama 7B Instruct Hfwon 3 · lost 0
Qwen2.5 0.5B Instructwon 3 · lost 0
Qwen2 7B Instructwon 3 · lost 0
Jamba Mini 1.7won 3 · lost 0
Gemma 2B Itwon 3 · lost 0
Magistral Smallwon 3 · lost 0
Aya 23 8Bwon 3 · lost 0
Laguna S 2.1won 3 · lost 0
Falcon H1 0.5B Instructwon 3 · lost 0
Llama 3.1 8B Instructwon 3 · lost 0
Grok Build 0.1won 3 · lost 0
Qwen1.5 14B Chatwon 3 · lost 0
DeepSeek R1 Distill Qwen 7Bwon 3 · lost 0
DeepSeek R1 Distill Llama 8Bwon 3 · lost 0
H2o Danube3 4B Chatwon 3 · lost 0
Zephyr 7B Betawon 3 · lost 0
Ling 3.0 Flash VLwon 3 · lost 0
Mimo V2.6 Flashwon 3 · lost 0
Magistral Mediumwon 3 · lost 0
Gemini 3.6 Flashwon 3 · lost 0
Jamba 1.6 Miniwon 3 · lost 0
Command R7Bwon 3 · lost 0
DeepSeek V4 Flashwon 3 · lost 0
Command Awon 3 · lost 0
Jamba Large 1.7won 3 · lost 0
NexusRaven V2 13Bwon 3 · lost 0
Ling 3.0 Flashwon 3 · lost 0
Qwen3 8Bwon 3 · lost 0
Open Mixtral 8x7bwon 3 · lost 0
Jamba 1.5 Largewon 3 · lost 0
GLM-5.2won 3 · lost 0
Gemini 3.5 Flashwon 3 · lost 0
Granite 3.2 8B Instructwon 3 · lost 0
Mistral 7B Instruct V0.2won 3 · lost 0
Nemotron 3 Ultra 550B A55Bwon 3 · lost 0
GLM-4.6won 3 · lost 0
Grok 4.3won 3 · lost 0
Phi-4 Mini Instructwon 3 · lost 0
Mimo V2.6 Pro Ultraspeedwon 3 · lost 0
Jamba 1.5 Miniwon 3 · lost 0
Qwen2.5 3B Instructwon 3 · lost 0
Aya 23 35Bwon 3 · lost 0
NVIDIA Nemotron 3 Super 120B A12Bwon 3 · lost 0
MiniMax M3won 3 · lost 0
Granite 3.0 1B A400M Instructwon 3 · lost 0
DeepSeek V3p1won 3 · lost 0
Gemma 4 E2Bwon 3 · lost 0
QwQ 32Bwon 3 · lost 0
Jamba 1.6 Largewon 3 · lost 0
Yi 1.5 9B Chatwon 3 · lost 0
Gemma 4 E4Bwon 3 · lost 0
Jamba Instructwon 3 · lost 0
SeaLLMs V3 7B Chatwon 3 · lost 0
Qwen2.5 7B Instruct 1Mwon 3 · lost 0
Mistral Smallwon 3 · lost 0
Kimi K2.5won 3 · lost 0
DeepSeek Reasonerwon 3 · lost 0
Qwen2.5 1.5B Instructwon 3 · lost 0
Mistral Small 24B Instructwon 3 · lost 0
DeepSeek V3.2won 3 · lost 0
Granite 3.1 2B Instructwon 3 · lost 0
WizardLM 2.8x22Bwon 3 · lost 0
Llama 3.2 3B Instructwon 3 · lost 0
Kimi K2.7 Codewon 3 · lost 0
Gemma 4 12Bwon 3 · lost 0
Hunyuan A13B Instructwon 3 · lost 0
Granite 3.2 2B Instructwon 3 · lost 0
Fuguwon 3 · lost 0
Llama 3.3 Instruct Turbo (70B)won 3 · lost 0
Granite 3.0 3B A800M Instructwon 3 · lost 0
Mistral Instruct v0.1 (7B)won 4 · lost 0
DBRX Instructwon 4 · lost 0
IBM Granite 4.0 Micro (with guardian)won 4 · lost 0
OLMoE 1B-7B Instructwon 3 · lost 1
OLMo 2 13B Instructwon 3 · lost 1
Mistral Instruct v0.3 (7B)won 3 · lost 1
Command Rwon 3 · lost 1
Mixtral (8x7B)won 3 · lost 1
OLMo 2 7B Instructwon 3 · lost 1
Grok 4won 3 · lost 1
GPT-3.5 Turbowon 3 · lost 1
Llama 3.1 Instruct Turbo (70B)won 3 · lost 1
Command R Pluswon 3 · lost 1
DeepSeek LLM 67B Chatwon 3 · lost 1
Marin 8B Instructwon 3 · lost 1
DeepSeek V3won 3 · lost 1
Palmyra Medwon 3 · lost 1
Llama 4 Scout Instructwon 3 · lost 1
IBM Granite 4.0 Microwon 3 · lost 1
Gemini 2.0 Flash Litewon 3 · lost 1
RakutenAI 7B Chatwon 2 · lost 1
PowerMoE 3Bwon 2 · lost 1
Phi 2won 2 · lost 1
PowerLM 3Bwon 2 · lost 1
SmolLM2 360M Instructwon 2 · lost 1
SEA LION V1 7B ITwon 2 · lost 1
LFM2 700Mwon 2 · lost 1
SmolLM 360M Instructwon 2 · lost 1
Gemma 2 9B Itwon 2 · lost 1
Open Codestral Mambawon 2 · lost 1
LFM2 1.2Bwon 2 · lost 1
GLM-4.5won 2 · lost 1
Kimi K2.6won 2 · lost 1
Sarvam 1won 2 · lost 1
Qwen3.7 Maxwon 2 · lost 1
Komodo 7B Basewon 2 · lost 1
Gemma 2 2B Itwon 2 · lost 1
Mimo V2.5 Prowon 2 · lost 1
Gemma 3 4B Itwon 2 · lost 1
Gemma 4 31Bwon 2 · lost 1
Starling Lm 7B Betawon 2 · lost 1
Open Mixtral 8x22bwon 2 · lost 1
Granite 3.0 2B Instructwon 2 · lost 1
Gemma2 9B Cpt Sea Lionv3 Instructwon 2 · lost 1
Reflection Llama 3.1 70Bwon 2 · lost 1
Apertus 8B Instructwon 2 · lost 1
Gemma 4 26B A4Bwon 2 · lost 1
DeepSeek Chatwon 2 · lost 1
DeepSeek V4 Prowon 2 · lost 1
Qwen2.5 32B Instructwon 2 · lost 1
LongWriter Glm4 9Bwon 2 · lost 1
Qwen2.5 14B Instruct 1Mwon 2 · lost 1
Qwen2 VL 72Bwon 2 · lost 1
GLM-5 Turbowon 2 · lost 1
Granite 3.1 3B A800M Instructwon 2 · lost 1
Phi 3 Medium 4k Instructwon 2 · lost 1
Granite 3.1 8B Instructwon 2 · lost 1
Sarvam 30Bwon 2 · lost 1
Qwen2.5 14B Instructwon 2 · lost 1
Phi 3 Medium 128k Instructwon 2 · lost 1
Gemma 3 27B Itwon 2 · lost 1
Qwen2 57B A14B Instructwon 2 · lost 1
Smaug Llama 3 70B Instructwon 2 · lost 1
Llama 3 70B Chat HFwon 2 · lost 1
Phi 3 Small 128k Instructwon 2 · lost 1
Granite 3.0 8B Instructwon 2 · lost 1
Mimo V2.6 Prowon 2 · lost 1
Gemma 3 12B Itwon 2 · lost 1
SeaLLM 7B V2won 2 · lost 1
Apertus 70B Instructwon 2 · lost 1
Granite 4.0 H Microwon 2 · lost 1
Granite 4.0 H Tinywon 2 · lost 1
Phi 3 Small 8k Instructwon 2 · lost 1
Phi 3 Mini 128k Instructwon 2 · lost 1
Phi 3.5 Mini Instructwon 2 · lost 1
K2 Chatwon 2 · lost 1
Llama 3.1 Tulu 3 8B SFTwon 2 · lost 1
Phi 3 Mini 4k Instructwon 2 · lost 1
Qwen3.8 Maxwon 2 · lost 1
Llama 3.1 Tulu 3 8B DPOwon 2 · lost 1
GLM-5.1won 2 · lost 1
Llama 3.1 Tulu 3 8Bwon 2 · lost 1
Qwen Maxwon 2 · lost 1
GPT-4won 2 · lost 1
Phi 3.5 MoE Instructwon 2 · lost 1
Nova Microwon 2 · lost 1
Llama 3 8B Chat Hfwon 2 · lost 1
GPT-5.6 Lunawon 2 · lost 1
C4ai Aya Expanse 32Bwon 2 · lost 1
C4ai Aya Expanse 8Bwon 2 · lost 1
Internlm2 Chat 20Bwon 2 · lost 1
Qwen2.5 72B Instructwon 2 · lost 1
Llama 2 70B Chat Hfwon 2 · lost 1
Nova Premierwon 2 · lost 1
Granite 4.0 H Smallwon 2 · lost 1
Llama 2 7B Chat Hfwon 2 · lost 1
Grok 3won 2 · lost 2
DeepSeek R1won 2 · lost 2
Mixtral Instruct (8x22B)won 2 · lost 2
Mistral Small 3won 2 · lost 2
DeepSeek R1 (hide reasoning)won 2 · lost 2
IBM Granite 3.3 8B Instructwon 2 · lost 2
Llama 3.1 Instruct Turbo (8B)won 2 · lost 2
Olmo 2 32B Instructwon 2 · lost 2
Gemini 1.5 Prowon 2 · lost 2
Llama 3.1 Instruct Turbo (405B)won 2 · lost 2
Qwen1.5 72B Chatwon 2 · lost 2
Gemini 1.5 Flashwon 2 · lost 2
Llama 3 8B Instructwon 2 · lost 2
Gemini 2.0 Flashwon 2 · lost 2
Llama 4 Maverick Instructwon 2 · lost 2
Llama 3 70B Instructwon 2 · lost 2
GPT-4.1 Nanowon 2 · lost 2
GPT-4o Miniwon 2 · lost 2
GPT-4.1 Miniwon 2 · lost 2
Qwen2 72B Instructwon 2 · lost 2
Dolphin 2.5 Mixtral 8x7bwon 1 · lost 2
GLM-5.3 Flashwon 1 · lost 2
Llama 3.2 1B Instructwon 1 · lost 2
Qwen3 Vl 235B A22B Instructwon 1 · lost 2
GPT-5.6 Astrawon 1 · lost 2
Claude Opus 4.6won 1 · lost 2
Claude Sonnet 4.6won 1 · lost 2
Kimi K2 Thinkingwon 1 · lost 2
GPT OSS Safeguard 20Bwon 1 · lost 2
Gemini 3.8 Flashwon 1 · lost 2
GPT-6 Solwon 1 · lost 2
Mistral Largewon 1 · lost 2
Nova Prowon 1 · lost 2
GPT-5.6 Solwon 1 · lost 2
Krutrim 2 Instructwon 1 · lost 2
Qwen3.5 35B A3Bwon 1 · lost 2
EAI SageAlign 8B LoRAwon 1 · lost 2
Qwen3.5 27Bwon 1 · lost 2
GPT-6 Lunawon 1 · lost 2
Qwen3.5 122B A10Bwon 1 · lost 2
Nova Litewon 1 · lost 2
Llama 3.2 Vision (11B)won 1 · lost 2
Llama 3.1 Nemotron 70B Instruct HFwon 1 · lost 2
Titan Text Express V1won 1 · lost 2
Llama 2 13B Chat Hfwon 1 · lost 2
Fugu Ultra V2won 1 · lost 2
Llama 3 8B Instruct RRwon 1 · lost 2
Grok 4.5won 1 · lost 2
GPT-5.5won 1 · lost 2
Sarvam 105Bwon 1 · lost 2
GPT-5.4won 1 · lost 2
Llama 3 8B Instruct MopeyMulewon 1 · lost 2
Llama 3.2 90B Vision Instruct Turbowon 1 · lost 2
GPT-5.6 Terrawon 1 · lost 2
Cogito 671B V2 P1won 1 · lost 2
Claude Opus 4.7won 1 · lost 2
Gemma 2 27B Itwon 1 · lost 2
Qwen3.5 397B A17Bwon 1 · lost 2
Meta SecAlign 8Bwon 1 · lost 2
Mistral NeMo Minitron 8B Instructwon 1 · lost 2
Claude Opus 5won 1 · lost 2
Grok 4.6won 1 · lost 2
Phi 4won 1 · lost 2
MiniMax M2won 1 · lost 2
Grok 4.7won 1 · lost 2
Gemini 2.5 Prowon 1 · lost 3
Gemini 2.5 Flashwon 1 · lost 3
Gemini 3 Prowon 1 · lost 3
Qwen 2.5 Instruct Turbo (7B)won 1 · lost 3
Claude 3 Haikuwon 1 · lost 3
Gemini 2.0 Prowon 1 · lost 3
IBM Granite 4.0 Smallwon 1 · lost 3
Qwen3 235B A22B Throughputwon 1 · lost 3
GLM-4.5 Airwon 1 · lost 3
Grok 3 Miniwon 1 · lost 3
Palmyra X5won 1 · lost 3
Palmyra Xwon 1 · lost 3
Qwen3 235B A22B Instructwon 1 · lost 3
Palmyra Finwon 1 · lost 3
Gemini 2.5 Flash Litewon 1 · lost 3
Qwen3 Next 80B A3B Thinkingwon 1 · lost 3
Claude 3.7 Sonnetwon 1 · lost 3
GPT-4owon 1 · lost 3
Claude 3 Sonnetwon 1 · lost 3
GPT-4.1won 1 · lost 3
Claude Opus 4won 1 · lost 3
Qwen 2.5 Instruct Turbo (72B)won 1 · lost 3
Claude Sonnet 4won 1 · lost 3
EAI Llama 3.1 8B Instruct SageAlignwon 0 · lost 3
Foundation Sec 8B Instructwon 0 · lost 3
Apriel 1.5 15B Thinkerwon 0 · lost 3
GPT-5.2won 0 · lost 3
Claude 3.5 Haikuwon 0 · lost 3
Muse Spark 1.1won 0 · lost 3
Claude Opus 4.8won 0 · lost 3
Inklingwon 0 · lost 3
Hy3won 0 · lost 3
MiniMax M2.7won 0 · lost 3
Claude Opus 5.5won 0 · lost 3
Claude Sonnet 5won 0 · lost 3
GPT OSS 20Bwon 0 · lost 4
Kimi K2won 0 · lost 4
O4 Miniwon 0 · lost 4
GPT OSS 120Bwon 0 · lost 4
GPT-4.5won 0 · lost 4
O3 Miniwon 0 · lost 4
Claude Haiku 4.5won 0 · lost 4
Claude Sonnet 4.5won 0 · lost 4
O1 Miniwon 0 · lost 4
GPT-4 Turbowon 0 · lost 4
GPT-5.1won 0 · lost 4
O3won 0 · lost 4
O1won 0 · lost 4
GPT-5 Miniwon 0 · lost 4
GPT-5won 0 · lost 4
GPT-5 Nanowon 0 · lost 4
Claude 3 Opuswon 0 · lost 4
Claude 3.5 Sonnetwon 0 · lost 4

§ 3 · Sources

Where the numbers come from

1 publication, 5 figures. Every one links to the page it was read from.

HELM Safety ↗5 measuresread 2026-09-27
HELM Safety · HarmBench 86.4%HELM Safety · SimpleSafetyTests 100%HELM Safety · Anthropic Red Team 99.9%HELM Safety · BBQ 90%HELM Safety · XSTest 80.6%

Badge

PublicAI Index badge for IBM Granite 4.0 Small (with guardian)[![PublicAI Index](https://publicai.io/model-index/badge?model=ibm-granite-4-0-small-with-guardian)](https://publicai.io/model-index/m/ibm-granite-4-0-small-with-guardian)