✓ Availablesync
DeepSeek: DeepSeek V4.1 Flash
deepseek/deepseek-v4.1-flashchatvisionreasoningstructured outputfunction calling
1,048,576 context tokens
$0.30 / $1.20 per 1M tokens
DeepSeek / Model DeveloperDeepSeek AI Models
on SandBase
Language models for reasoning and general text tasks.
Choose the DeepSeek model capability that fits your next task.
Language
8model & capability records
8 / 0 / 0 / 0language · image · video · audio
8 Sync0 async records
Catalog snapshot · Oct 10, 2026 · Task variants, not independent foundation models.
Choose a DeepSeek Model
Balanced Agent WorkDeepSeek: DeepSeek V4.1 Flash ↗$300000/M inputHigh-Volume Text on a BudgetDeepSeek: DeepSeek V4 Flash Vision Exp (Experimental) ↗$300000/M inputAdvanced ReasoningDeepSeek: DeepSeek V4 Pro ↗$1320000/M inputImage Generation & EditingDeepSeek: DeepSeek V4 Flash ↗$300000/M inputCinematic VideoDeepSeek: DeepSeek V3 ↗$320000/M inputSpeech & MusicDeepSeek: DeepSeek V4 Flash 0731 ↗$300000/M input
Recommendations are task-oriented starting points, not performance rankings. Confirm live pricing and schema.
EXPLORE THE CATALOG
DeepSeek Models & Capabilities
Model IDs, task variants, and pricing units.
All in one place.
8 model & capability recordsCatalog snapshot · Oct 10, 2026
Token prices are per million input/output tokens; request prices are base prices. Context is shown only when meaningful. Confirm pricing, availability, and request fields on the live model page.
Gemini
Language, multimodal understanding, reasoning, and tool use.
✓ Availablesync
DeepSeek: DeepSeek V4 Flash Vision Exp (Experimental)
deepseek/deepseek-v4-flash-vision-expchatvisionreasoningstructured outputfunction calling
1,048,576 context tokens
$0.30 / $1.20 per 1M tokens✓ Availablesync
DeepSeek: DeepSeek V4 Pro
deepseek/deepseek-v4-prochatreasoningstructured outputfunction calling
1,048,576 context tokens
$1.32 / $3.96 per 1M tokens✓ Availablesync
DeepSeek: DeepSeek V4 Flash
deepseek/deepseek-v4-flashchatreasoningstructured outputfunction calling
1,048,576 context tokens
$0.30 / $1.20 per 1M tokens✓ Availablesync
DeepSeek: DeepSeek V3
deepseek/deepseek-chatchatstructured outputfunction calling
163,840 context tokens
$0.32 / $0.89 per 1M tokens✓ Availablesync
DeepSeek: DeepSeek V4 Flash 0731
deepseek/deepseek-v4-flash-0731chatreasoningstructured outputfunction calling
1,048,576 context tokens
$0.30 / $1.20 per 1M tokens✓ Availablesync
DeepSeek: DeepSeek V4 Pro 0813
deepseek/deepseek-v4-pro-0813chatreasoningstructured outputfunction calling
1,048,576 context tokens
$1.32 / $3.96 per 1M tokens✓ Availablesync
DeepSeek: DeepSeek V3.2
deepseek/deepseek-v3.2chatreasoningstructured outputfunction calling
131,072 context tokens
$0.25 / $0.38 per 1M tokensUse DeepSeek Models in an Agent Workflow
Keep your model selection separate from the Agent. Check tool support and schema compatibility when swapping models.