Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

Llama 3 70B Instruct

Meta · 70B

Strongest in Toxicity avoidance (#2 of 272), weakest in Human preference (#239 of 342). Above par in 6 of 11 scopes. Among the models it meets almost everywhere, it finishes behind Qwen3 Next 80B A3B Thinking and GPT-5 and ahead of Grok 4 and GPT-4.1 Nano.

§ 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.

Safety52.4#116/337
Toxicity avoidance59#2/272
Safe-prompt compliance55.9#29/82
Jailbreak resistance59.8#51/272
Fairness50.1#114/300
Harm refusal51.7#130/300
Secure code44#200/274
Core abilities44#161/204
General intelligence41.4#164/204
Human preference45.2#239/342
Human preference45.2#239/342

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

OLMoE 1B-7B Instructwon 7 · lost 0
Mistral Instruct v0.1 (7B)won 7 · lost 0
Mistral Instruct v0.3 (7B)won 7 · lost 0
Llama 3.1 Instruct Turbo (405B)won 7 · lost 0
Magistral Smallwon 8 · lost 0
Jamba Mini 1.7won 8 · lost 0
Smollm2 1.7B Instructwon 8 · lost 0
Zephyr 7B Betawon 8 · lost 0
Gemma 7B Itwon 8 · lost 0
Mistral 7B Instruct V0.2won 8 · lost 0
Nous Hermes 2 Mixtral 8x7b Dpowon 8 · lost 0
Granite 3.1 2B Instructwon 8 · lost 0
Llama 3.1 8B Instructwon 8 · lost 0
Mistral Small 24B Instructwon 8 · lost 0
Command R Pluswon 9 · lost 0
DeepSeek LLM 67B Chatwon 8 · lost 1
DBRX Instructwon 8 · lost 1
GPT-3.5 Turbowon 8 · lost 1
Nemotron 3 Ultra 550B A55Bwon 7 · lost 1
Grok Build 0.1won 7 · lost 1
Gemma 2B Itwon 7 · lost 1
Granite 3.0 2B Instructwon 7 · lost 1
Llama 3.2 3B Instructwon 7 · lost 1
Starling Lm 7B Betawon 7 · lost 1
Qwen1.5 14B Chatwon 7 · lost 1
Phi 3 Medium 4k Instructwon 7 · lost 1
Granite 3.1 8B Instructwon 7 · lost 1
Jamba 1.5 Miniwon 7 · lost 1
Gemma 2 9B Itwon 7 · lost 1
OLMo 2 7B Instructwon 6 · lost 1
Llama 3.1 Instruct Turbo (70B)won 6 · lost 1
IBM Granite 4.0 Microwon 6 · lost 1
Llama 3 8B Instructwon 7 · lost 2
Kimi K2.7 Codewon 6 · lost 2
Jamba Large 1.7won 6 · lost 2
Llama 3.2 1B Instructwon 6 · lost 2
Granite 3.0 8B Instructwon 6 · lost 2
Jamba 1.5 Largewon 6 · lost 2
Claude 3 Haikuwon 8 · lost 3
Llama 3.1 Instruct Turbo (8B)won 5 · lost 2
Qwen 2.5 Instruct Turbo (7B)won 5 · lost 2
Gemma 4 12Bwon 5 · lost 2
Gemma 3 4B Itwon 7 · lost 3
Magistral Mediumwon 7 · lost 3
Command Awon 7 · lost 3
Grok 4.3won 7 · lost 3
Gemini 1.5 Flashwon 6 · lost 3
Gemini 1.5 Prowon 6 · lost 3
Gemini 2.0 Flash Litewon 6 · lost 3
Llama 4 Scout Instructwon 7 · lost 4
Ling 3.0 Flash VLwon 5 · lost 3
Phi 3 Mini 128k Instructwon 5 · lost 3
Phi 3 Mini 4k Instructwon 5 · lost 3
Phi 3 Small 8k Instructwon 5 · lost 3
Gemma 2 2B Itwon 5 · lost 3
Nova Microwon 5 · lost 3
Nova Litewon 5 · lost 3
C4ai Aya Expanse 32Bwon 5 · lost 3
GPT-4won 5 · lost 3
QwQ 32Bwon 5 · lost 3
Gemma 3 12B Itwon 5 · lost 3
NVIDIA Nemotron 3 Super 120B A12Bwon 5 · lost 3
Qwen3 30B A3B Instructwon 5 · lost 3
Grok 4.1 Fastwon 5 · lost 3
Gemma 4 26B A4Bwon 5 · lost 3
Mimo V2.6 Flashwon 5 · lost 3
Mistral Smallwon 6 · lost 4
Grok 3won 6 · lost 4
GPT-4o Miniwon 6 · lost 5
GPT-4.1 Nanowon 6 · lost 5
Grok 4won 6 · lost 5
Kimi K2 Thinkingwon 4 · lost 4
Llama 3.1 Tulu 3 8Bwon 4 · lost 4
C4ai Aya Expanse 8Bwon 4 · lost 4
Phi 4won 4 · lost 4
Nova Prowon 4 · lost 4
Qwen3 Vl 235B A22B Instructwon 4 · lost 4
Gemma 3 27B Itwon 5 · lost 5
GLM-4.6won 5 · lost 5
DeepSeek V4 Flashwon 5 · lost 5
MiniMax M3won 5 · lost 5
Kimi K2.5won 5 · lost 5
Gemma 4 31Bwon 5 · lost 5
Kimi K2.6won 5 · lost 5
DeepSeek V4 Prowon 5 · lost 5
GLM-5.2won 5 · lost 5
Gemini 3.5 Flashwon 5 · lost 5
Mimo V2.6 Prowon 5 · lost 5
Gemini 3.6 Flashwon 5 · lost 5
Llama 4 Maverick Instructwon 5 · lost 6
DeepSeek R1won 5 · lost 6
Claude 3.7 Sonnetwon 4 · lost 5
GLM-4.5won 4 · lost 6
DeepSeek V3.2won 4 · lost 6
GPT-6 Solwon 4 · lost 6
Mimo V2.5 Prowon 4 · lost 6
GLM-5.3 Flashwon 4 · lost 6
Qwen3.7 Maxwon 4 · lost 6
Gemini 3.8 Flashwon 4 · lost 6
Qwen2.5 72B Instructwon 3 · lost 5
Mistral Largewon 3 · lost 5
Qwen Maxwon 3 · lost 5
Qwen3.5 27Bwon 3 · lost 5
GPT OSS 20Bwon 4 · lost 7
Gemini 2.0 Flashwon 4 · lost 7
DeepSeek V3won 4 · lost 7
Gemini 2.5 Flashwon 4 · lost 7
Gemini 2.5 Prowon 4 · lost 7
Qwen2 72B Instructwon 3 · lost 6
Qwen3.5 122B A10Bwon 3 · lost 7
GPT-6 Lunawon 3 · lost 7
GPT-5.6 Lunawon 3 · lost 7
GPT-5.4won 3 · lost 7
Qwen3.8 Maxwon 3 · lost 7
Qwen 2.5 Instruct Turbo (72B)won 2 · lost 5
GPT OSS 120Bwon 3 · lost 8
GPT-4.1 Miniwon 3 · lost 8
Kimi K2won 3 · lost 8
Claude Sonnet 4.5won 3 · lost 8
Gemma 2 27B Itwon 2 · lost 6
Claude 3.5 Haikuwon 2 · lost 6
Qwen3.5 35B A3Bwon 2 · lost 6
GLM-5.1won 2 · lost 6
Claude Sonnet 4.6won 2 · lost 6
Gemini 3 Prowon 2 · lost 6
Claude 3 Sonnetwon 2 · lost 7
Claude 3 Opuswon 2 · lost 7
GPT-4owon 2 · lost 7
O3 Miniwon 2 · lost 7
Claude 3.5 Sonnetwon 2 · lost 7
GPT-5.1won 2 · lost 7
MiniMax M2won 2 · lost 8
Qwen3.5 397B A17Bwon 2 · lost 8
Grok 4.6won 2 · lost 8
Grok 4.5won 2 · lost 8
GPT-5.6 Terrawon 2 · lost 8
GPT-5.5won 2 · lost 8
GPT-5.6 Solwon 2 · lost 8
Claude Opus 5won 2 · lost 8
O4 Miniwon 2 · lost 9
Claude Sonnet 4won 2 · lost 9
Claude Haiku 4.5won 2 · lost 9
GPT-4.1won 2 · lost 9
Qwen3 235B A22B Instructwon 2 · lost 9
GLM-4.5 Airwon 1 · lost 7
Gemini 2.5 Flash Litewon 1 · lost 7
MiniMax M2.7won 1 · lost 7
Claude Opus 4.6won 1 · lost 7
GPT-4 Turbowon 1 · lost 8
O1 Miniwon 1 · lost 8
O1won 1 · lost 8
GPT-5.2won 1 · lost 9
Grok 4.7won 1 · lost 9
Inklingwon 1 · lost 9
Hy3won 1 · lost 9
Claude Sonnet 5won 1 · lost 9
Claude Opus 4.8won 1 · lost 9
Muse Spark 1.1won 1 · lost 9
Claude Opus 4.7won 1 · lost 9
Claude Opus 5.5won 1 · lost 9
GPT-5 Nanowon 1 · lost 10
GPT-5 Miniwon 1 · lost 10
Claude Opus 4won 1 · lost 10
O3won 1 · lost 10
GPT-5won 1 · lost 10
Qwen3 Next 80B A3B Thinkingwon 0 · lost 8

§ 3 · Sources

Where the numbers come from

4 publications, 13 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26
LMArena Text 1277
Kagi LLM Benchmark ↗1 measureread 2026-09-26
Kagi LLM Benchmark 35.1%
HELM Safety ↗5 measuresread 2026-09-26
HELM Safety · HarmBench 64%HELM Safety · SimpleSafetyTests 99%HELM Safety · Anthropic Red Team 96.7%HELM Safety · BBQ 91%HELM Safety · XSTest 96.8%
Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-09-26
Enkrypt · Jailbreak risk 4.5%Enkrypt · Harmful content risk 12.2%Enkrypt · CBRN risk 13.7%Enkrypt · Toxicity risk 0%Enkrypt · Bias risk 80.4%Enkrypt · Insecure code risk 45.8%

Badge

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