DeepSeek V3.1
- Context
- 164K
- Released
- 1y ago
- Parameters
- 671B
- Input / 1M tokens
- $0.25
- Output / 1M tokens
- $0.95
- Cached input / 1M tokens
- $0.13
Sources:OpenRouter
Benchmarks
DeepSWE is the primary coding benchmark where an exact model evaluation exists. Other sourced evaluations appear below; missing scores are not zero.
Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.
No benchmark results have been matched to this model yet.
Specifications
- Developer
- DeepSeek
- Model family
- DeepSeek
- Type
- Language model
- Predecessor
- —
- Successor
- —
- Released
- August 21, 2025
- Model ID
deepseek/deepseek-chat-v3.1- Openness
- Open weights
- License
- —
- Parameters
- 671B
- Input modalities
- text
- Output modalities
- text
- Modality
- text->text
- Native context
- 160K
- Tokenizer
- DeepSeek
- Max output tokens
- 33K
- Knowledge cutoff
- 2025-03-31
- Hugging Face
- deepseek-ai/DeepSeek-V3.1
- Source freshness
- Endpoints synced Oct 15, 2026, 12:18 PM UTC
Open-weight builds
Ollama variants
Every published Ollama tag matched to this model, including parameter editions, instruction/base variants, and quantized builds.
| Variant | Parameters | Type | Quantization | Size | Context | Input |
|---|---|---|---|---|---|---|
| deepseek-v3.1:671b-fp16 | 671B | — | FP16 | 1.3 TB | 160K | text |
| deepseek-v3.1:671b-terminus-fp16 | 671B | — | FP16 | 1.3 TB | 160K | text |
| deepseek-v3.1:latest | 671B | — | Q4_K_M | 404 GB | 160K | text |
| deepseek-v3.1:671b-q8_0 | 671B | — | Q8_0 | 713 GB | 160K | text |
| deepseek-v3.1:671b-terminus-q8_0 | 671B | — | Q8_0 | 713 GB | 160K | text |
Infrastructure
Provider availability
| Provider | Endpoint | Context | Input /M tokens | Output /M tokens | Cached input /M tokens | Cache write /M tokens | Uptime 1d | Data handling |
|---|---|---|---|---|---|---|---|---|
| deepinfra/fp4 | 164K | $0.25 | $0.95 | $0.13 | — | 100.00% | ZDR: YesTraining: Unknown Prompt logging: Unknown Retention: Unknown Checked 2026-10-15 | |
| siliconflow/fp8 | 164K | $0.27 | $1.00 | — | — | 97.50% | ZDR: YesTraining: Unknown Prompt logging: Unknown Retention: Unknown Checked 2026-10-15 | |
| coreweave/fp8 | 161K | $0.55 | $1.65 | $0.55 | — | 100.00% | ZDR: YesTraining: Unknown Prompt logging: Unknown Retention: Unknown Checked 2026-10-15 | |
| mara | 131K | $0.60 | $1.70 | — | — | 98.68% | ZDR: YesTraining: Unknown Prompt logging: Unknown Retention: Unknown Checked 2026-10-15 | |
| sambanova/fp8 | 131K | $0.65 | $1.50 | — | — | 97.16% | ZDR: YesTraining: Unknown Prompt logging: Unknown Retention: Unknown Checked 2026-10-15 |
Provider data:OpenRouter
Deployment graph
Can I run this?
Vertical Tabs only shows compatibility when it has structured evidence tying this exact model to hardware, runtime, quantization, memory use, and observed performance.