The LLM benchmark aggregator.

Smaug Llama 3 70B Instruct

Abacus.AI · 70B

Strongest in Toxicity avoidance (#6 of 272), weakest in Secure code (#233 of 274). Above par in 4 of 6 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 4.6 and Claude Sonnet 5 and ahead of Phi 3.5 MoE Instruct and Qwen Max.

§ 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.

Safety51.7−9.8#140/3361/3

Toxicity avoidance59leads#6/2721/1

Jailbreak resistance58.6−6.3#64/2721/1

Harm refusal53−8.6#110/2991/2

Fairness50−20#117/2991/2

Secure code36.5−30#233/2741/1

§ 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 Smaug Llama 3 70B Instruct, left for the other.

Mistral Instruct v0.1 (7B)won 6 · lost 0

OLMo 7B Instruct Hfwon 6 · lost 0

SmolLM 1.7B Instructwon 6 · lost 0

H2o Danube3 500M Chatwon 6 · lost 0

Smollm2 1.7B Instructwon 6 · lost 0

LFM2 350Mwon 6 · lost 0

RakutenAI 7B Instructwon 6 · lost 0

Nous Hermes 2 Mixtral 8x7b Dpowon 6 · lost 0

Mistral Instruct v0.3 (7B)won 6 · lost 0

Magistral Smallwon 6 · lost 0

Aya 23 8Bwon 6 · lost 0

DeepSeek R1 Distill Qwen 7Bwon 6 · lost 0

DeepSeek R1 Distill Llama 8Bwon 6 · lost 0

H2o Danube3 4B Chatwon 6 · lost 0

Command R Pluswon 6 · lost 0

Granite 3.2 8B Instructwon 6 · lost 0

Mistral 7B Instruct V0.2won 6 · lost 0

WizardLM 2.8x22Bwon 6 · lost 0

Llama 4 Scout Instructwon 6 · lost 0

Llama 3.3 Instruct Turbo (70B)won 6 · lost 0

Reflection Llama 3.1 70Bwon 6 · lost 0

OLMoE 1B-7B Instructwon 5 · lost 1

RakutenAI 7B Chatwon 5 · lost 1

DBRX Instructwon 5 · lost 1

Phi 2won 5 · lost 1

Qwen2.5 7B Instructwon 5 · lost 1

Gemma 7B Itwon 5 · lost 1

Grok 4.1 Fastwon 5 · lost 1

Qwen3 30B A3B Instructwon 5 · lost 1

Laguna Xs 2.1won 5 · lost 1

Falcon Mamba 7B Instructwon 5 · lost 1

GLM-4 9B Chatwon 5 · lost 1

SmolLM2 360M Instructwon 5 · lost 1

CodeLlama 7B Instruct Hfwon 5 · lost 1

Qwen2.5 0.5B Instructwon 5 · lost 1

Qwen2 7B Instructwon 5 · lost 1

Jamba Mini 1.7won 5 · lost 1

Grok 4won 5 · lost 1

Gemma 2B Itwon 5 · lost 1

Laguna S 2.1won 5 · lost 1

Falcon H1 0.5B Instructwon 5 · lost 1

SEA LION V1 7B ITwon 5 · lost 1

GPT-3.5 Turbowon 5 · lost 1

Llama 3.1 8B Instructwon 5 · lost 1

Grok Build 0.1won 5 · lost 1

Qwen1.5 14B Chatwon 5 · lost 1

DeepSeek R1won 5 · lost 1

Zephyr 7B Betawon 5 · lost 1

Ling 3.0 Flash VLwon 5 · lost 1

Mimo V2.6 Flashwon 5 · lost 1

Magistral Mediumwon 5 · lost 1

Gemini 3.6 Flashwon 5 · lost 1

Llama 3.1 Instruct Turbo (70B)won 5 · lost 1

Jamba 1.6 Miniwon 5 · lost 1

Command R7Bwon 5 · lost 1

LFM2 700Mwon 5 · lost 1

DeepSeek V4 Flashwon 5 · lost 1

Command Awon 5 · lost 1

SmolLM 360M Instructwon 5 · lost 1

NexusRaven V2 13Bwon 5 · lost 1

Ling 3.0 Flashwon 5 · lost 1

Qwen3 8Bwon 5 · lost 1

Open Mixtral 8x7bwon 5 · lost 1

Jamba 1.5 Largewon 5 · lost 1

Gemma 2 9B Itwon 5 · lost 1

GLM-5.2won 5 · lost 1

DeepSeek LLM 67B Chatwon 5 · lost 1

Gemini 3.5 Flashwon 5 · lost 1

Nemotron 3 Ultra 550B A55Bwon 5 · lost 1

GLM-4.6won 5 · lost 1

Grok 4.3won 5 · lost 1

Phi-4 Mini Instructwon 5 · lost 1

Open Codestral Mambawon 5 · lost 1

Mimo V2.6 Pro Ultraspeedwon 5 · lost 1

Jamba 1.5 Miniwon 5 · lost 1

Qwen2.5 3B Instructwon 5 · lost 1

Aya 23 35Bwon 5 · lost 1

NVIDIA Nemotron 3 Super 120B A12Bwon 5 · lost 1

MiniMax M3won 5 · lost 1

Granite 3.0 1B A400M Instructwon 5 · lost 1

DeepSeek V3p1won 5 · lost 1

Gemma 4 E2Bwon 5 · lost 1

QwQ 32Bwon 5 · lost 1

Jamba 1.6 Largewon 5 · lost 1

Yi 1.5 9B Chatwon 5 · lost 1

Gemma 4 E4Bwon 5 · lost 1

Jamba Instructwon 5 · lost 1

Sarvam 1won 5 · lost 1

SeaLLMs V3 7B Chatwon 5 · lost 1

Qwen2.5 7B Instruct 1Mwon 5 · lost 1

Mistral Smallwon 5 · lost 1

Kimi K2.5won 5 · lost 1

DeepSeek Reasonerwon 5 · lost 1

Qwen2.5 1.5B Instructwon 5 · lost 1

Mistral Small 24B Instructwon 5 · lost 1

Granite 3.1 2B Instructwon 5 · lost 1

Llama 3.2 3B Instructwon 5 · lost 1

Kimi K2.7 Codewon 5 · lost 1

Gemma 4 12Bwon 5 · lost 1

Hunyuan A13B Instructwon 5 · lost 1

Granite 3.2 2B Instructwon 5 · lost 1

Fuguwon 5 · lost 1

Gemini 2.0 Flash Litewon 5 · lost 1

Granite 3.0 3B A800M Instructwon 5 · lost 1

Gemma 3 4B Itwon 5 · lost 1

Gemma 4 31Bwon 5 · lost 1

Starling Lm 7B Betawon 5 · lost 1

Open Mixtral 8x22bwon 5 · lost 1

Granite 3.0 2B Instructwon 5 · lost 1

Llama 3.1 Instruct Turbo (405B)won 5 · lost 1

Gemma2 9B Cpt Sea Lionv3 Instructwon 5 · lost 1

Apertus 8B Instructwon 5 · lost 1

Gemma 4 26B A4Bwon 5 · lost 1

DeepSeek Chatwon 5 · lost 1

DeepSeek V4 Prowon 5 · lost 1

Qwen2.5 32B Instructwon 5 · lost 1

LongWriter Glm4 9Bwon 5 · lost 1

Qwen2.5 14B Instruct 1Mwon 5 · lost 1

Gemini 1.5 Flashwon 5 · lost 1

Qwen2 VL 72Bwon 5 · lost 1

GLM-5 Turbowon 5 · lost 1

Phi 3 Medium 4k Instructwon 5 · lost 1

Qwen2.5 14B Instructwon 5 · lost 1

Phi 3 Medium 128k Instructwon 5 · lost 1

Gemma 3 27B Itwon 5 · lost 1

PowerMoE 3Bwon 4 · lost 2

PowerLM 3Bwon 4 · lost 2

OLMo 2 7B Instructwon 4 · lost 2

Grok 3won 4 · lost 2

Jamba Large 1.7won 4 · lost 2

DeepSeek V3won 4 · lost 2

LFM2 1.2Bwon 4 · lost 2

GLM-4.5won 4 · lost 2

Kimi K2.6won 4 · lost 2

Qwen3.7 Maxwon 4 · lost 2

Komodo 7B Basewon 4 · lost 2

DeepSeek V3.2won 4 · lost 2

Gemma 2 2B Itwon 4 · lost 2

IBM Granite 4.0 Microwon 4 · lost 2

Mimo V2.5 Prowon 4 · lost 2

Llama 3.1 Instruct Turbo (8B)won 4 · lost 2

Dolphin 2.5 Mixtral 8x7bwon 4 · lost 2

Gemini 1.5 Prowon 4 · lost 2

Gemini 2.5 Prowon 4 · lost 2

GLM-5.3 Flashwon 4 · lost 2

Llama 3.2 1B Instructwon 4 · lost 2

Gemini 2.5 Flashwon 4 · lost 2

Llama 3 8B Instructwon 4 · lost 2

Qwen 2.5 Instruct Turbo (7B)won 4 · lost 2

Sarvam 30Bwon 4 · lost 2

Qwen3 Vl 235B A22B Instructwon 4 · lost 2

Gemini 2.0 Flashwon 4 · lost 2

Qwen2 57B A14B Instructwon 4 · lost 2

Phi 3 Small 128k Instructwon 4 · lost 2

Granite 3.0 8B Instructwon 4 · lost 2

Mimo V2.6 Prowon 4 · lost 2

Gemma 3 12B Itwon 4 · lost 2

SeaLLM 7B V2won 4 · lost 2

Phi 3 Small 8k Instructwon 4 · lost 2

Qwen3.8 Maxwon 4 · lost 2

Qwen Maxwon 4 · lost 2

Phi 3.5 MoE Instructwon 4 · lost 2

GPT OSS 20Bwon 3 · lost 3

Granite 3.1 3B A800M Instructwon 3 · lost 3

Granite 3.1 8B Instructwon 3 · lost 3

Claude 3 Haikuwon 3 · lost 3

GPT-5.6 Astrawon 3 · lost 3

Kimi K2 Thinkingwon 3 · lost 3

GPT OSS Safeguard 20Bwon 3 · lost 3

Gemini 3.8 Flashwon 3 · lost 3

Apertus 70B Instructwon 3 · lost 3

GPT-6 Solwon 3 · lost 3

Granite 4.0 H Microwon 3 · lost 3

Granite 4.0 H Tinywon 3 · lost 3

Phi 3 Mini 128k Instructwon 3 · lost 3

Mistral Largewon 3 · lost 3

K2 Chatwon 3 · lost 3

Llama 3.1 Tulu 3 8B SFTwon 3 · lost 3

GPT-4.1 Nanowon 3 · lost 3

Nova Prowon 3 · lost 3

GPT-4o Miniwon 3 · lost 3

GLM-5.1won 3 · lost 3

Qwen3.5 35B A3Bwon 3 · lost 3

EAI SageAlign 8B LoRAwon 3 · lost 3

GPT-4.1 Miniwon 3 · lost 3

Qwen3.5 27Bwon 3 · lost 3

GPT-6 Lunawon 3 · lost 3

Qwen3.5 122B A10Bwon 3 · lost 3

Nova Microwon 3 · lost 3

GPT-5.6 Lunawon 3 · lost 3

Nova Litewon 3 · lost 3

Fugu Ultra V2won 3 · lost 3

Granite 4.0 H Smallwon 3 · lost 3

Grok 4.5won 3 · lost 3

Sarvam 105Bwon 3 · lost 3

GPT-5.4won 3 · lost 3

Qwen3.5 397B A17Bwon 3 · lost 3

Mistral NeMo Minitron 8B Instructwon 3 · lost 3

Claude Opus 5won 3 · lost 3

Llama 3 70B Chat HFwon 2 · lost 4

Kimi K2won 2 · lost 4

EAI Llama 3.1 8B Instruct SageAlignwon 2 · lost 4

Llama 4 Maverick Instructwon 2 · lost 4

Phi 3.5 Mini Instructwon 2 · lost 4

Phi 3 Mini 4k Instructwon 2 · lost 4

GPT-5.6 Solwon 2 · lost 4

Llama 3.1 Tulu 3 8B DPOwon 2 · lost 4

Krutrim 2 Instructwon 2 · lost 4

Llama 3.1 Tulu 3 8Bwon 2 · lost 4

GPT-4won 2 · lost 4

Llama 3 8B Chat Hfwon 2 · lost 4

C4ai Aya Expanse 32Bwon 2 · lost 4

C4ai Aya Expanse 8Bwon 2 · lost 4

Qwen3 235B A22B Instructwon 2 · lost 4

Llama 3.2 Vision (11B)won 2 · lost 4

GPT OSS 120Bwon 2 · lost 4

Llama 3.1 Nemotron 70B Instruct HFwon 2 · lost 4

Titan Text Express V1won 2 · lost 4

Internlm2 Chat 20Bwon 2 · lost 4

Qwen2.5 72B Instructwon 2 · lost 4

Llama 2 70B Chat Hfwon 2 · lost 4

Llama 2 13B Chat Hfwon 2 · lost 4

Llama 3 8B Instruct RRwon 2 · lost 4

Nova Premierwon 2 · lost 4

Llama 2 7B Chat Hfwon 2 · lost 4

GPT-5.5won 2 · lost 4

Claude 3.7 Sonnetwon 2 · lost 4

Llama 3 8B Instruct MopeyMulewon 2 · lost 4

GPT-4.1won 2 · lost 4

GPT-5.6 Terrawon 2 · lost 4

Cogito 671B V2 P1won 2 · lost 4

Qwen 2.5 Instruct Turbo (72B)won 2 · lost 4

Meta SecAlign 8Bwon 2 · lost 4

Apriel 1.5 15B Thinkerwon 2 · lost 4

GPT-5.1won 2 · lost 4

Grok 4.6won 2 · lost 4

Phi 4won 2 · lost 4

MiniMax M2won 2 · lost 4

Claude Sonnet 4.6won 1 · lost 3

Llama 3 70B Instructwon 1 · lost 5

O4 Miniwon 1 · lost 5

GPT-4owon 1 · lost 5

Qwen2 72B Instructwon 1 · lost 5

O3 Miniwon 1 · lost 5

Claude Haiku 4.5won 1 · lost 5

Claude 3 Sonnetwon 1 · lost 5

Llama 3.2 90B Vision Instruct Turbowon 1 · lost 5

Claude Opus 4won 1 · lost 5

Claude Opus 4.7won 1 · lost 5

Claude Sonnet 4.5won 1 · lost 5

Foundation Sec 8B Instructwon 1 · lost 5

Gemma 2 27B Itwon 1 · lost 5

O1 Miniwon 1 · lost 5

Claude Sonnet 4won 1 · lost 5

GPT-4 Turbowon 1 · lost 5

O3won 1 · lost 5

GPT-5.2won 1 · lost 5

O1won 1 · lost 5

GPT-5 Miniwon 1 · lost 5

Claude 3.5 Haikuwon 1 · lost 5

GPT-5won 1 · lost 5

GPT-5 Nanowon 1 · lost 5

Claude 3 Opuswon 1 · lost 5

Muse Spark 1.1won 1 · lost 5

Claude Opus 4.8won 1 · lost 5

Inklingwon 1 · lost 5

Grok 4.7won 1 · lost 5

Claude 3.5 Sonnetwon 1 · lost 5

Hy3won 1 · lost 5

MiniMax M2.7won 1 · lost 5

Claude Opus 5.5won 1 · lost 5

Claude Sonnet 5won 1 · lost 5

Claude Opus 4.6won 0 · lost 4

§ 3 · Sources

Where the numbers come from

1 publication, 6 figures. Every one links to the page it was read from.

Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-09-27

Enkrypt · Jailbreak risk 5.5%Enkrypt · Harmful content risk 30.6%Enkrypt · CBRN risk 8.5%Enkrypt · Toxicity risk 0%Enkrypt · Bias risk 77.8%Enkrypt · Insecure code risk 60.9%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=smaug-llama-3-70b-instruct)](https://publicai.io/model-index/m/smaug-llama-3-70b-instruct)