Primary use stated or inferred cautiously from official documentation.
D
DEEPSEEK / deepseek-reasoner
DeepSeek R1
Open reasoning model and distilled family, relevant for tracking the evolution of verifiable reasoning.
01 / SUMMARY
The essential
Maximum input capacity when published by the source.
Documented maximum generation limit.
Verified availability channels.
Primary functional family and verified specialties.
Declared terms for API access, model weights, or self-hosting.
02 / TECHNICAL SHEET
Limits and integration
| API ID | deepseek-reasoner |
|---|---|
| Model type | Text and reasoning · Code |
| Access model | API and open weights |
| License | MIT |
| Deployment | API, local, or private cloud |
| Release | 20 ENE 2025 |
| Knowledge cutoff | Not published |
| Entrance | Text |
| Exit | Text |
| Context window | 64,000 tokens in the original API |
| maximum output | Not published |
| Reasoning | Explicit reasoning |
| Published tools | |
| Structured Outings | Not applicable or not published |
| Batch processing | Not published |
| Prompt cache | Not published |
| Fine-tuning | Downloadable weights and distilled models |
| Verified platforms | DeepSeek API · Hugging Face |
Mathematics, code, research, and study of open reasoning models.
It is an earlier generation; compare latency and reasoning length with V3.2.
03 / CAPABILITIES
What it can do
Orientation
Mathematics, code, research, and study of open reasoning models.
Context
64,000 tokens in the original API · Not published
Tools and integration
Not published
Access
DeepSeek API and weights
Input modalities
Text
04 / PRICES
Documented cost
| Concept | Worth | Unit/condition |
|---|---|---|
| Standard input | Not published | Standard API rate |
| Cached input | Not published | Reading reused prefixes |
| Cache write or storage | Not published | The condition varies by provider |
| Standard output | Not published | May include reasoning tokens |
| Batch input | Not published | Asynchronous processing |
| Batch output | Not published | Asynchronous processing |
Consult the primary source: the billing unit depends on the model type and access channel.
i Prices change and may depend on level, region or context length. Check the source before making a decision.
05 / EVALUATIONS
How to read the results
“
A public benchmark provides guidance, but does not replace an evaluation with your data, tools, budget, and error tolerance.
| Benchmark | Result | Metric | Source |
|---|---|---|---|
| AIME 2024 | 79,8 % | Published Pass@1 | View source ↗ |
Inferama only highlights a “best result” when the metric, test set, configuration, and date allow for an equivalent comparison. The supplier's figures are presented as claims from its own source.
06 / VERSIONS
Chronology
DeepSeek R1
Version added to Inferama's verified catalog.
ANALYSIS
Related analysis
DeepSeek R1 in Production: How to Test the Prompt Template Without Breaking Your Application
Official prompting recommendations for DeepSeek R1 are hypotheses worth validating in each application. This protocol compares instruction placement and the “<think>” prefix while measuring both answer quality and integration failures.
27 Sep 2026 ↗
DeepSeek R1 vs. DeepSeek V3.2: how to choose between explicit reasoning and efficiency in a reviewable workflow
A useful comparison between DeepSeek R1 and DeepSeek V3.2 does not begin with a published benchmark, but with an organization’s own task, a frozen corpus, and an explicit definition of material error. This protocol makes it possible to measure whether R1’s additional deliberation reduces errors and human review, or whether V3.2 reaches the same threshold with lower latency, consumption, and complexity.
22 Sep 2026 ↗
AIME 2024: What an Accuracy Percentage Across 30 Problems Measures—and Why Two Apparently Identical Results May Not Be Comparable
An AIME 2024 figure can look straightforward while summarizing very different protocols. This guide explains what the dataset represents, how results change with the number of attempts, answer-selection methods, tools, and evaluation harnesses, and what information to request before comparing models.
22 Sep 2026 ↗07 / SOURCES