Back to all LLMs

DeepSeek R1

163,840 token context with transparent <think> delimiters showing reasoning over retrieved documents. MIT license enables fine-tuning on domain-specific retrieval tasks and full model customization. If you want to compare the best LLMs for your data, try Agentset.

Leaderboard Rank
#14
of 16
ELO Rating
1325
#14
Win Rate
22.5%
#15
Latency
18272ms
#14

Model Information

Provider
DeepSeek
License
Open Source
Input Price per 1M
$0.30
Output Price per 1M
$1.20
Context Window
164K
Release Date
2025-01-20
Model Name
deepseek-r1
Total Evaluations
1350

Performance Record

Wins304 (22.5%)
Losses885 (65.6%)
Ties161 (11.9%)
Wins
Losses
Ties

LLMs Are Just One Piece of RAG

Agentset gives you a managed RAG pipeline with the top-ranked models and best practices baked in. No infrastructure to maintain, no LLM orchestration to manage.

Trusted by teams building production RAG applications

5M+
Documents
1,500+
Teams
99.9%
Uptime

Performance Overview

ELO ratings by dataset

DeepSeek R1's ELO performance varies across different benchmark datasets, showing its strengths in specific domains.

DeepSeek R1 - ELO by Dataset

Detailed Metrics

Dataset breakdown

Performance metrics across different benchmark datasets, including accuracy and latency percentiles.

SciFact

ELO 149326.0% WR117W-241L-92T

Quality Metrics

Correctness
4.97
Faithfulness
4.97
Grounding
4.97
Relevance
5.00
Completeness
4.80
Overall
4.94

Latency Distribution

Mean
14826ms
Min
7765ms
Max
33129ms

PG

ELO 141524.2% WR109W-320L-21T

Quality Metrics

Correctness
4.87
Faithfulness
4.87
Grounding
4.87
Relevance
4.93
Completeness
4.60
Overall
4.83

Latency Distribution

Mean
23334ms
Min
12280ms
Max
85633ms

MSMARCO

ELO 106817.3% WR78W-324L-48T

Quality Metrics

Correctness
4.67
Faithfulness
4.70
Grounding
4.67
Relevance
4.83
Completeness
4.57
Overall
4.69

Latency Distribution

Mean
16654ms
Min
9675ms
Max
31255ms

Build RAG in Minutes, Not Months

Agentset gives you a complete RAG API with top-ranked LLMs and smart retrieval built in. Upload your data, call the API, and get grounded answers from day one.

import { Agentset } from "agentset";

const agentset = new Agentset();
const ns = agentset.namespace("ns_1234");

const results = await ns.search(
  "What is multi-head attention?"
);

for (const result of results) {
  console.log(result.text);
}

Compare Models

See how it stacks up

Compare DeepSeek R1 with other top llms to understand the differences in performance, accuracy, and latency.