deepseek-llm-7B-chat Benchmark Results

deepseek-llm-7B-chat: benchmark results

DeepSeek LLM 7B chat checkpoint. Provider: DeepSeek. Released 2023-11-29. Access: Open.

Unified ELO 1409 ± 1, rank #1100 of 1392 rated models, from 17 benchmark results.

Benchmark	Score	Metric	Percentile
Open LLM Leaderboard - MuSR	19.21	Score	93
ChineseSafe Benchmark	71.63	Accuracy (%)	79.6
Open Chinese LLM - HellaSwag	62.57	Accuracy (%)	67.7
Open Chinese LLM - WinoGrande	64.17	Accuracy (%)	65
Open LLM Leaderboard - IFEval	41.71	Score	42.5
Open Chinese LLM - ARC Challenge	49.66	Accuracy (%)	40.2
Open Chinese LLM - TruthfulQA MC	51.61	Accuracy (%)	29.8
MedCTA	11	Outcome Accuracy (self-reported)	29.4
Open Chinese LLM - GSM8K	31.54	Accuracy (%)	25.5
Open Chinese LLM Leaderboard	51.57	Average Score (%)	24
Open LLM Leaderboard - MMLU-Pro	12.59	Score	22.4
Open LLM Leaderboard - BBH	11.26	Score	20.8

Interactive version: theaggregate.ai/model?slug=deepseek-llm-7b-chat · How It Works · Data refreshed daily, snapshot 2026-09-05.