DeepSeek Prover Models by DeepSeek
2 models2024–2025Up to 160K ctxFrom $0.56/1M input
About
DeepSeek Prover is a family of 2 AI models by DeepSeek, released between 2024 and 2025.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
2 in view
DeepSeek Prover V2Current
Use when the workload needs reasoning, 160K context, and 671B parameters.
2025-01reasoning160K context671B parameters
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| DeepSeek Prover V2 | Use when the workload needs reasoning, 160K context, and 671B parameters. | 2025-01 | reasoning160K context671B parameters | Current |
| DeepSeek Prover | Use when provider availability and model metadata match the workload. | 2024-01 | — | Current |
Release Timeline
2 release groups2025-01
1 current
DeepSeek Prover V2
Currentreasoning160K context671B parameters
2024-01
1 current
Current
Specifications(2 models)
| Model | Released | Context | Parameters | Reasoning |
|---|---|---|---|---|
| DeepSeek Prover V2 | 2025-01 | 160K | 671B | Yes |
| DeepSeek Prover | 2024-01 | — | — | No |
Available From(1 provider)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| DeepSeek Prover V2 | Fireworks AI | $0.56 | $1.68 | Serverless |
Frequently Asked Questions
- What is DeepSeek Prover used for?
- DeepSeek Prover is used for reasoning and math-heavy prompts. The family description and listed model capabilities point to those workloads as the best fit.
- How does DeepSeek Prover compare to Janus?
- DeepSeek Prover by DeepSeek is strongest where you need reasoning, while Janus by DeepSeek is the closest related family to check for image generation. DeepSeek Prover has 2 listed variants and reaches up to 160K context, so compare the specs and pricing tables before choosing a production model.
- Which DeepSeek Prover model should I use?
- For the lowest listed input price, start with DeepSeek Prover V2 through Fireworks AI at $0.56/1M input tokens. For the most capable/latest local choice, evaluate DeepSeek Prover V2 with 160K context and reasoning.
