DeepSeek-R1

DeepSeek-R1 (deepseek via Azure) costs $1.35 per 1M input tokens and $5.40 per 1M output tokens, as last published by Azure.

DeepSeek-R1 has a context window of 163,840 tokens (the maximum combined input and output it can handle in a single request).

DeepSeek-R1 sits in the large size tier. Size tiers run from micro (small enough to run on a device) through small, medium and large to extra large (frontier flagships), and are assigned from the model's family, naming and limits rather than its price.

DeepSeek-R1 can generate up to 163,840 tokens in a single response. Reasoning (thinking) tokens count toward this limit.

DeepSeek-R1's training data runs through July 2024, so it may not know about later events. Coolhand records the cutoff at month precision, since providers often publish only the month.

DeepSeek-R1 accepts text as input and produces text as output, as recorded in the Coolhand catalog.

Yes — the weights of DeepSeek-R1 can be downloaded under the MIT license. Check the license terms for any limits on commercial use.

Yes — deepseek deprecated DeepSeek-R1 on August 13, 2026. Listed as deprecated on models.dev as of 2026-08-13.

deepseek retired DeepSeek-R1 on August 13, 2026.

Since DeepSeek-R1 is deprecated, deepseek customers are moving to DeepSeek-R1, DeepSeek-V3.2, and DeepSeek-V3.1.

Call GET https://coolhandlabs.com/api/v2/inference_apis?q[source_api_eq]=azure&q[model_eq]=deepseek-r1 — free, no API key required, and scoped to just this model since source_api and model together are unique. See https://coolhandlabs.com/docs for the full REST API reference.

Coolhand also tracks DeepSeek-R1, DeepSeek-V3.2, DeepSeek-V3.1, DeepSeek V4 Flash, DeepSeek V4 Flash Free, and DeepSeek V3.1 from deepseek. Some may already be deprecated, so check each model's status alongside its current pricing.