R1
DeepSeekhuggingface.co/deepseek-ai/DeepSeek-R1
DeepSeek R1 is here: Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active in an inference pass....
Updated 39 min ago · first seen 11 Sept 2026
model_01M294WWKV997B1F6QH020KYNQ
- Parameters
- 684.5B
- T2 · 5 h ago
- Context
- 64K tokens
- T2 · 5 h ago
- Released
- 20 Jan 2025
- T2 · 5 h ago
- License
- mit
- T2 · 5 h ago
- Knowledge cutoff
- Jul 2024
- T2 · 5 h ago
Specification
- Release date
- 20 Jan 2025
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Openness
- open-weights
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- License
- mit
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Architecture
- DeepseekV3ForCausalLM
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Parameters
- 684.5B
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Context window
- 64K tokens
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Max output
- 16K tokens
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Knowledge cutoff
- Jul 2024
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Modalities
- text
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Input modalities
- text
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Output modalities
- text
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Tokenizer
- DeepSeek
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Quantization format
- fp8
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- File size
- 688.6 GB
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Hugging Face repo
- deepseek-ai/DeepSeek-R1
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Pipeline tag
- text-generation
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Model card
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Downloads
- 670,913
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 39 min agomedium
- Likes
- 13,619
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 39 min agomedium
- Gated
- No
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Hf inference providers
- novita
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Instruct type
- deepseek-r1
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Is quantized
- Yes
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Last modified
- 2025-03-27T04:01:59+00:00
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Library name
- transformers
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Downloads all time
- 48,545,119
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 39 min agomedium
- Model type
- deepseek_v3
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Openrouter id
- deepseek/deepseek-r1
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Supported parameters
- frequency_penalty, include_reasoning, max_tokens, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_p
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Tags
- transformers, safetensors, deepseek_v3, text-generation, custom_code, fp8
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
- Weights dtype
- BF16, F32, F8_E4M3
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 5 h agomedium
Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →
Provenance
Attributed facts
35
Source tiers
T235
Freshest observation
39 min ago
Conflicts
None
Modalities
- Modalities
- text
- Input
- text
- Output
- text
Capabilities
Tool calling
Yes
OpenRouter public model & pricing listing · T2
Structured output
Yes
OpenRouter public model & pricing listing · T2
Reasoning
Yes
OpenRouter public model & pricing listing · T2
Vision
Unavailable
Audio
Unavailable
Fine-tuning available
Unavailable
- Context window
- 64K tokens
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Max output
- 16K tokens
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Knowledge cutoff
- Jul 2024
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
- Tokenizer
- DeepSeek
Source:OpenRouter public model & pricing listingT2observed 5 h agomedium
| Benchmark | Score | Metric | Config | Evaluated | Source |
|---|---|---|---|---|---|
| Aider polyglot | 56.9% | pass_rate_2 | date=2025-01-20 · command=aider --model deepseek/deepseek-reasoner · dirname=2025-01-20-19-11-38--ds-turns-upd-cur-msgs-fix-with-summarizer · versions=0.71.2.dev | 20 Jan 2025 | aider.chatT2 |
| Aider polyglot | 96.9% | percent_cases_well_formed | date=2025-01-20 · command=aider --model deepseek/deepseek-reasoner · dirname=2025-01-20-19-11-38--ds-turns-upd-cur-msgs-fix-with-summarizer · versions=0.71.2.dev | 20 Jan 2025 | aider.chatT2 |
Scores are reported as published, with their evaluation configuration (harness, prompting, judge). Results with different configs are not directly comparable — see methodology.
Current prices
| Provider | Input / 1M | Output / 1M | Cached in | Batch in / out | Context | Observed | Source |
|---|---|---|---|---|---|---|---|
| DeepSeek APIdeepseek/deepseek-r1 | $0.70 | $2.5 | — | — | 64K | 1 h ago | openrouter.aiT2 |
USD per 1M tokens as published by each provider (USD). Rows are append-only: every change is kept in the history below.
Available through 1
Price history
Input price · USD / 1M tokens
Output price · USD / 1M tokens
- DeepSeek API
Memory need = bytes per parameter (4-bit ≈ 0.5 × 1.15 overhead, 8-bit 1.0, fp16 2.0) + a KV-cache allowance. Not a measurement.
| Hardware | Quantization | Memory | Est. need | Fits |
|---|---|---|---|---|
| NVIDIA DGX B200 | 4bit | 1,440 GB | 394.1 GB | Yes |
| Apple M3 Ultra | 4bit | — | 394.1 GB | Yes |
| Mac Studio (Apple M5 Ultra) | 4bit | — | 394.1 GB | Yes |
| AMD Instinct MI325X | 4bit | 256 GB | 394.1 GB | No |
| AMD Instinct MI300X | 4bit | 192 GB | 394.1 GB | No |
| Apple M2 Ultra | 4bit | — | 394.1 GB | No |
| NVIDIA B200 | 4bit | 180 GB | 394.1 GB | No |
| NVIDIA H200 | 4bit | 141 GB | 394.1 GB | No |
| NVIDIA H200 NVL | 4bit | 141 GB | 394.1 GB | No |
| Apple M1 Ultra | 4bit | — | 394.1 GB | No |
| Apple M3 Max | 4bit | — | 394.1 GB | No |
| Apple M4 Max | 4bit | — | 394.1 GB | No |
| Mac Studio (Apple M5 Max) | 4bit | — | 394.1 GB | No |
| MacBook Pro (Apple M5 Max) | 4bit | — | 394.1 GB | No |
| NVIDIA DGX Spark | 4bit | 128 GB | 394.1 GB | No |
| Apple M2 Max | 4bit | — | 394.1 GB | No |
| NVIDIA H100 NVL | 4bit | 94 GB | 394.1 GB | No |
| NVIDIA A100 80GB | 4bit | 80 GB | 394.1 GB | No |
| NVIDIA H100 SXM | 4bit | 80 GB | 394.1 GB | No |
| Apple M1 Max | 4bit | — | 394.1 GB | No |
| Apple M4 Pro | 4bit | — | 394.1 GB | No |
| Mac mini (Apple M5 Pro) | 4bit | — | 394.1 GB | No |
| MacBook Pro (Apple M5 Pro) | 4bit | — | 394.1 GB | No |
| Apple M3 Pro | 4bit | — | 394.1 GB | No |
| Apple M1 Pro | 4bit | — | 394.1 GB | No |
| Apple M2 Pro | 4bit | — | 394.1 GB | No |
| Apple M4 | 4bit | — | 394.1 GB | No |
| iMac (Apple M4) | 4bit | — | 394.1 GB | No |
| Mac mini (Apple M6) | 4bit | — | 394.1 GB | No |
| MacBook Air (Apple M5) | 4bit | — | 394.1 GB | No |
| MacBook Pro (Apple M5) | 4bit | — | 394.1 GB | No |
| NVIDIA GeForce RTX 5090 | 4bit | 32 GB | 394.1 GB | No |
| Apple M2 | 4bit | — | 394.1 GB | No |
| Apple M3 | 4bit | — | 394.1 GB | No |
| NVIDIA GeForce RTX 3090 | 4bit | 24 GB | 394.1 GB | No |
| NVIDIA GeForce RTX 4090 | 4bit | 24 GB | 394.1 GB | No |
| Apple M1 | 4bit | — | 394.1 GB | No |
Ancestors 0
None recorded.
Descendants 0
None recorded.
Papers 1
- arXiv:2501.12948Active35
Repositories 0
No repositories linked yet.
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history · Architecture
Architecturearchitecture1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| DeepseekV3ForCausalLM | → current | current | Hugging Face Hub (public pages, model cards, papers)T2 | medium | deterministic |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
DeepSeek API lists R1 at $0.7 in / $2.5 out per 1M tokens
openrouter
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/deepseek-ai/DeepSeek-R1/raw/main/README.md | model_card | T2· Quality secondary | 3 min ago | 1 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/deepseek-ai/DeepSeek-R1 | model_page | T2· Quality secondary | 39 min ago | 3 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/amd/DeepSeek-R1-MXFP4 | model_page | T2· Quality secondary | 49 min ago | 4 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/models?author=deepseek-ai&p=0&sort=downloads | listing | T2· Quality secondary | 53 min ago | 4 |
| OpenRouter public model & pricing listing | openrouter.ai/api/v1/models | catalogue | T2· Quality secondary | 1 h ago | 4 |
| Aider polyglot leaderboard (data file on GitHub) | raw.githubusercontent.com/Aider-AI/aider/main/aider/website/_data/polyglot_leaderboard.yml | leaderboard | T2· Quality secondary | 4 h ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.