Shaberi3 ベンチマーク

LLMパフォーマンス比較ダッシュボード

Upload CSV File

あなたのベンチマーク結果を可視化したい場合は、shaberi/results/totals.csvをアップロードしてください

Select Models

chatgpt-4o-latest

claude-3-5-sonnet-20241022

o1-mini-2024-09-12

gpt-4o-2024-05-13

gemini-1.5-pro-exp-0827

EZO-Qwen2.5-72B-Instruct-Q8_0

Athene-V2-Agent:Q4_K_M

Athene-V2-Chat:Q4_K_M

claude-3-5-sonnet-20240620

claude-3-opus-20240229

EZO-Qwen2.5-72B-Instruct-Q4_K_M

qwen2.5:72b-instruct-q8_0

gemini-1.5-pro-002

Llama-3.1-70B-EZO-1.1-it-Q8_0

gpt-4o-2024-08-06

Llama-3.1-Nemotron-70B-Instruct-Q4_K_L

qwen2.5:32b-instruct-q8_0

claude-3-5-haiku-20241022

gpt-4o-mini-2024-07-18

gemini-1.5-pro

EZO-Qwen2.5-32B-Instruct-Q4_K_M

EZO-Qwen2.5-32B-Instruct-Q8_0

gemini-1.5-flash-002

Athene-70B-Q8_0

gemma-2-9B-it-SimPO-Q8_0

gemini-1.5-flash-exp-0827

EZO-Humanities-9B-gemma-2-it-f16

gemma-2-9B-it-SimPO-f16

Mistral-Large-Instruct-2407-Q4_K_M

qwen2:72b-instruct-q4_K_M

c4ai-aya-expanse-32b

gemma-2-27b-it.f16.Q8_0

EZO-Common-9B-gemma-2-it-f16

magnum-v2-72b-Q8_0

qwen2.5:14b-instruct-fp16

gemini-1.5-flash

SuperNova-Medius-f16

deepseek-chat

gemma-2-9b-it.f16.Q8_0

gemini-1.5-flash-8b

gemini-1.5-flash-8b-exp-0827

qwen2.5-coder:32b-instruct-q8_0

claude-3-haiku-20240307

Llama-3.1-Swallow-70B-Instruct-v0.1-Q4_K_M

qwen2.5-coder:32b-instruct-q4_K_M

c4ai-command-r-plus-Q4_K_M

c4ai-aya-expanse-8b

Tiger-Gemma-9B-v3-f16

qwen2.5:7b-instruct-fp16

gpt-3.5-turbo-0125

Llama-3.1-Swallow-8B-Instruct-v0.1-f16

llama3.2-vision:90b-instruct-q4_K_M

Llama-3.1-Swallow-8B-Instruct-v0.2-Q8_0

command-r-plus:104b-08-2024-q4_K_M

mistral-small:22b-instruct-2409-q8_0

Llama-3.1-Swallow-8B-Instruct-v0.2-Q4_K_M

mixtral:8x22b-instruct-v0.1-q4_K_M

llama3.1:70b-instruct-q4_K_M

EZO-Common-T2-2B-gemma-2-it-f16

Ministral-8B-Instruct-2410-HF-f16

gemma-2-baku-2b-it-f16

EZO-gemma-2-2b-jpn-it-f16

ministral-8b-instruct-2410_q4km

EZO-gemma-2-2b-jpn-it-f16-mod

lfm-40b

llm-jp-3-13b-instruct-f16

gemma2:2b-instruct-fp16

gemma-2-baku-2b-it-f16-mod

qwen2.5-coder:7b-instruct-fp16

gemma-2-2b-jpn-it-f16

Mistral-NeMo-Minitron-8B-Instruct-f16

mixtral:8x7b-instruct-v0.1-q8_0

llm-jp-3-3.7b-instruct-EZO-Humanities-f16

qwen2.5:3b-instruct-fp16

Borea-Phi-3.5-mini-Instruct-Common-f16

Borea-Phi-3.5-mini-Instruct-Jp-f16

gemma-2-2b-jpn-it-f16-mod

llm-jp-3-3.7b-instruct-EZO-Common-f16

Phi-3.5-mini-instruct-f16

llama3.1:8b-instruct-fp16

llm-jp-3-3.7b-instruct-f16

llama3.2-vision:11b-instruct-q4_K_M

granite3-dense:8b-instruct-fp16

llama3.2:3b-instruct-fp16

qwen2.5:1.5b-instruct-fp16

llm-jp-3-1.8b-instruct-f16

granite3-dense:2b-instruct-fp16

llm-jp-3-172b-beta1-instruct-Q3_K_S

raspberry-3B-f16

granite3-moe:3b-instruct-fp16

llama3.2:1b-instruct-fp16

qwen2.5:0.5b-instruct-fp16

granite3-moe:1b-instruct-fp16

SmolLM2-1.7B-Instruct:fp16

Select Metrics

Weighted Mean

ELYZA-tasks-100

MT-Bench

Tengu-Bench

Weighted Mean
ELYZA-tasks-100
MT-Bench
Tengu-Bench

7.5