Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

Llama 3.1 70B Instruct

Meta · 71B · open weights

Strongest in Safe-prompt compliance (#54 of 82), weakest in Safety (#275 of 331). Above par in 5 of 12 scopes. Among the models it meets almost everywhere, it finishes behind Claude Sonnet 4.5 and GPT-5.1 and ahead of Gemini 1.5 Flash and Gemma 2 9B It.

§ 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.

Professional42.8#163/187
Medical50.7#80/149
Finance33.3#157/161
Human preference46.5#229/346
Human preference46.5#229/346
Safety45.7#275/331
Safe-prompt compliance50.8#54/82
Jailbreak resistance55.4#111/271
Fairness50.1#114/296
Toxicity avoidance52.6#163/271
Harm refusal43.3#264/296
Secure code26#269/273

§ 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 3.1 70B Instruct, left for the other.

Mixtral 8x7b Instructwon 10 · lost 1
DBRX Instructwon 8 · lost 1
Gemma 2B Itwon 7 · lost 1
Smollm2 1.7B Instructwon 7 · lost 1
Nous Hermes 2 Mixtral 8x7b Dpowon 7 · lost 1
GPT-3.5 Turbowon 7 · lost 2
Zephyr 7B Betawon 6 · lost 2
Gemma 7B Itwon 6 · lost 2
Qwen1.5 14B Chatwon 6 · lost 2
Jamba 1.5 Miniwon 8 · lost 3
Jamba 1.6 Miniwon 6 · lost 3
DeepSeek LLM 67B Chatwon 6 · lost 3
Command R Pluswon 7 · lost 4
Jamba 1.5 Largewon 7 · lost 4
Magistral Mediumwon 5 · lost 3
Jamba 1.6 Largewon 5 · lost 4
Llama 3 8B Instructwon 5 · lost 4
Gemma 2 9B Itwon 5 · lost 4
Gemini 1.5 Flashwon 6 · lost 5
Granite 3.0 2B Instructwon 4 · lost 4
Llama 3.2 3B Instructwon 4 · lost 4
Phi 3 Small 8k Instructwon 4 · lost 4
Starling Lm 7B Betawon 4 · lost 4
Gemma 2 2B Itwon 4 · lost 4
Mistral Small 24B Instructwon 4 · lost 4
GPT-4won 4 · lost 4
Qwen3 30B A3B Instructwon 4 · lost 4
Llama 3.1 8B Instructwon 6 · lost 6
Grok 4won 6 · lost 6
Mixtral 8x22b Instructwon 5 · lost 6
Mistral Smallwon 5 · lost 6
Grok 4.1 Fastwon 5 · lost 6
Claude 3 Haikuwon 4 · lost 5
Gemini 2.0 Flash Litewon 4 · lost 5
Gemini 1.5 Prowon 5 · lost 7
Llama 3.2 1B Instructwon 3 · lost 5
Phi 3 Mini 128k Instructwon 3 · lost 5
Phi 3 Mini 4k Instructwon 3 · lost 5
Mistral 7B Instruct V0.2won 3 · lost 5
Granite 3.1 2B Instructwon 3 · lost 5
Granite 3.0 8B Instructwon 3 · lost 5
Phi 3 Medium 4k Instructwon 3 · lost 5
Granite 3.1 8B Instructwon 3 · lost 5
Llama 3.1 Tulu 3 8Bwon 3 · lost 5
C4ai Aya Expanse 8Bwon 3 · lost 5
Nova Microwon 3 · lost 5
Phi 4won 3 · lost 5
Nova Litewon 3 · lost 5
C4ai Aya Expanse 32Bwon 3 · lost 5
Nova Prowon 3 · lost 5
Gemma 3 4B Itwon 3 · lost 5
Command Awon 4 · lost 7
NVIDIA Nemotron 3 Ultra 550B A55Bwon 4 · lost 7
DeepSeek V4 Flashwon 4 · lost 7
Grok 4.3won 4 · lost 7
Gemini 3.6 Flashwon 4 · lost 7
Ling 3.0 Flashwon 3 · lost 6
Llama 3 70B Instructwon 3 · lost 6
Claude 3 Sonnetwon 3 · lost 6
GPT OSS 20Bwon 4 · lost 8
GPT-4.1 Nanowon 4 · lost 8
GPT-4.1 Miniwon 4 · lost 8
Grok 3won 4 · lost 8
DeepSeek V3.1won 3 · lost 7
Mistral Largewon 3 · lost 8
GLM-4.6won 3 · lost 8
Kimi K2.5won 3 · lost 8
Mimo V2.6 Flashwon 3 · lost 8
GPT-5.6 Lunawon 3 · lost 8
Kimi K2.6won 3 · lost 8
DeepSeek V4 Prowon 3 · lost 8
Qwen3.8 Maxwon 3 · lost 8
QwQ 32Bwon 2 · lost 6
NVIDIA Nemotron 3 Super 120B A12Bwon 2 · lost 6
Qwen3 Vl 235B A22B Instructwon 2 · lost 6
Gemma 4 26B A4Bwon 2 · lost 6
Llama 4 Scout Instructwon 3 · lost 9
Llama 4 Maverick Instructwon 3 · lost 9
GPT OSS 120Bwon 3 · lost 9
DeepSeek V3won 3 · lost 9
Kimi K2won 3 · lost 9
DeepSeek R1won 3 · lost 9
Qwen2 72B Instructwon 2 · lost 7
Gemma 2 27B Itwon 2 · lost 7
Gemma 4 31Bwon 2 · lost 8
DeepSeek V3.2won 2 · lost 9
Mistral Large 4won 2 · lost 9
MiniMax M3won 2 · lost 9
GPT-6 Lunawon 2 · lost 9
GPT-6 Solwon 2 · lost 9
GLM-5.1won 2 · lost 9
Grok 4.5won 2 · lost 9
GPT-5.6 Terrawon 2 · lost 9
Mimo V2.5 Prowon 2 · lost 9
Qwen3.7 Maxwon 2 · lost 9
GLM-5.2won 2 · lost 9
Gemini 3.5 Flashwon 2 · lost 9
GPT-5.5won 2 · lost 9
GPT-5.6 Solwon 2 · lost 9
GPT-4o Miniwon 2 · lost 10
Gemini 2.0 Flashwon 2 · lost 10
O3 Miniwon 2 · lost 10
O4 Miniwon 2 · lost 10
Gemini 2.5 Flashwon 2 · lost 10
Claude Haiku 4.5won 2 · lost 10
Llama 3.2 Vision (11B)won 1 · lost 7
Qwen 2.5 Instruct (7B)won 1 · lost 7
Qwen Maxwon 1 · lost 7
Gemma 3 12B Itwon 1 · lost 7
Mercury 2won 1 · lost 7
Gemma 3 27B Itwon 1 · lost 7
Qwen3.5 35B A3Bwon 1 · lost 7
Qwen3.5 27Bwon 1 · lost 7
Qwen3.5 122B A10Bwon 1 · lost 7
Qwen3.5 397B A17Bwon 1 · lost 7
Kimi K2 Thinkingwon 1 · lost 8
Claude 3 Opuswon 1 · lost 8
Gemini 2.5 Flash Litewon 1 · lost 8
Claude Sonnet 4.6won 1 · lost 8
Llama 3.3 70B Instructwon 1 · lost 9
Qwen2.5 72B Instructwon 1 · lost 10
GPT-4 Turbowon 1 · lost 10
GLM-4.5won 1 · lost 10
MiniMax M2.7won 1 · lost 10
GLM-5.3 Flashwon 1 · lost 10
GPT-5.4won 1 · lost 10
Mimo V2.6 Prowon 1 · lost 10
Claude Opus 5won 1 · lost 10
Gemini 3.8 Flashwon 1 · lost 10
Llama 3.1 405B Instructwon 1 · lost 11
GPT-5 Nanowon 1 · lost 11
Claude 3.5 Sonnetwon 1 · lost 11
Claude 3.7 Sonnetwon 1 · lost 11
GPT-5 Miniwon 1 · lost 11
Claude Sonnet 4won 1 · lost 11
GPT-4.1won 1 · lost 11
Qwen3 235B A22B Instructwon 1 · lost 11
O3won 1 · lost 11
Gemini 2.5 Prowon 1 · lost 11
Llama 3.2 Vision (90B)won 0 · lost 8
MiniMax M2won 0 · lost 8
Hy3won 0 · lost 8
Grok 3 Miniwon 0 · lost 9
Gemini 3 Prowon 0 · lost 9
Claude Opus 4.6won 0 · lost 9
Claude 3.5 Haikuwon 0 · lost 10
O1 Miniwon 0 · lost 11
GPT-5.2won 0 · lost 11
Inklingwon 0 · lost 11
Grok 4.7won 0 · lost 11
Grok 4.6won 0 · lost 11
Claude Sonnet 5won 0 · lost 11
Claude Opus 4.8won 0 · lost 11
Muse Spark 1.1won 0 · lost 11
Claude Opus 4.7won 0 · lost 11
Claude Opus 5.5won 0 · lost 11
GPT-4owon 0 · lost 12
O1won 0 · lost 12
Claude Opus 4won 0 · lost 12
GPT-5won 0 · lost 12
GPT-5.1won 0 · lost 12
Claude Sonnet 4.5won 0 · lost 12

§ 3 · Sources

Where the numbers come from

4 publications, 15 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-10-19
LMArena Text 1293
Vals.ai ↗3 measuresread 2026-10-19
Vals · CorpFin 38.85%Vals · TaxEval 56.17%Vals · MedQA 84.78%
HELM Safety ↗5 measuresread 2026-10-19
HELM Safety · HarmBench 46.9%HELM Safety · SimpleSafetyTests 92.5%HELM Safety · Anthropic Red Team 93.2%HELM Safety · BBQ 95.4%HELM Safety · XSTest 94.5%
Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-10-19
Enkrypt · Jailbreak risk 8.2%Enkrypt · Harmful content risk 37.2%Enkrypt · CBRN risk 13.5%Enkrypt · Toxicity risk 4.4%Enkrypt · Bias risk 86.3%Enkrypt · Insecure code risk 92.9%

Badge

PublicAI Index badge for Llama 3.1 70B Instruct[![PublicAI Index](https://publicai.io/model-index/badge?model=llama-3-1-70b-instruct)](https://publicai.io/model-index/m/llama-3-1-70b-instruct)