DeepSeek V3.1 Terminus
DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing issues reported by users, including language consistency and agent capabilities, further optimizing the model's…
DeepSeek-V3.1-TerminusReleased 22 Sep 2025 (Sep 2025)Pricing
List prices at DeepInfra, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $0.27 | $1.00 | |
| Cached input 20% off input | $0.216 | 20% off input |
What it costs a month
At DeepInfra list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$151
- RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$153
- Coding agent1,000 requests a day, 40K in / 2K out, cached$339
- Overnight batch20,000 requests a day, 4K in / 0.4K out$888
Where to run it
4 hosts serve DeepSeek V3.1 Terminus. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
DeepInfraInference DeepSeek-V3.1-Terminus | $0.27 / $1.00 | DeepSeek-V3.1-Terminus | |
GMI CloudInference DeepSeek-V3.1-Terminus | $0.27 / $1.00 | DeepSeek-V3.1-Terminus | |
Novita AIInference deepseek-v3.1-terminus | $0.27 / $1.00 | deepseek-v3.1-terminus | |
Fireworks AIInference deepseek-v3p1-terminus | $0.56 / $1.68 +86% vs headline | deepseek-v3p1-terminus |
Capabilities
Core
- Streaming
- Function calling
- JSON mode
- Structured outputs
Multimodal
- Vision
- Image input
- Audio
- Image generation
- Video
Advanced
- Reasoning
- Tool calling
- MCP
- Prompt caching
- Batch API
- Embeddings
- Takes in
- Text
- Produces
- Text
- Open weights
- License: DeepSeek License (MIT code)Runs in about 1642.9 GB of GPU memoryRuntimes: SGLang, TGI, vLLM
Cheaper alternatives
Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.
Same model, cheaper host
No host sells it for less than DeepInfra.
Smaller DeepSeek models
No cheaper DeepSeek model keeps its key capabilities.
Other frontier labs
- 56% less
- 56% less
- 42% less
Open-weight picks
- 1% less
- 3% less
- 4% less
- 13% less
Call it
Using DeepInfra's API.
from litellm import completion
response = completion(
model="deepinfra/DeepSeek-V3.1-Terminus",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does DeepSeek V3.1 Terminus cost?
On DeepInfra, DeepSeek V3.1 Terminus costs $0.27 per million input tokens and $1.00 per million output tokens. Cached input costs $0.216 per million tokens.
Which APIs offer DeepSeek V3.1 Terminus?
DeepSeek V3.1 Terminus runs on 4 APIs: DeepInfra, GMI Cloud, Novita AI and Fireworks AI.
What is DeepSeek V3.1 Terminus's context window?
164K tokens, with up to 164K tokens in one reply.
Is DeepSeek V3.1 Terminus being retired?
No retirement date is scheduled for it on any API we track.
What are cheaper alternatives to DeepSeek V3.1 Terminus?
Models that cost less and keep most of what it can do: Claude Haiku 5.5 on Anthropic (56% less), GPT-6 Luna on OpenAI (56% less) and GLM 4 6v on Novita AI (1% less). Savings are on a mix of 3 input tokens to 1 output token.
When was DeepSeek V3.1 Terminus released?
DeepSeek released it on 22 Sep 2025.
What is DeepSeek V3.1 Terminus's API model id?
Use DeepSeek-V3.1-Terminus with DeepInfra. Other hosts use their own ids: deepseek-v3.1-terminus on Novita AI, deepseek-v3p1-terminus on Fireworks AI.
Sources
Raw catalog entry for DeepSeek-V3.1-Terminus on DeepInfra
{
"max_input_tokens": 163840,
"max_output_tokens": 163840,
"input_cost_per_token": 2.7e-7,
"output_cost_per_token": 0.000001,
"cache_read_input_token_cost": 2.16e-7,
"litellm_provider": "deepinfra",
"mode": "chat",
"supports_tool_choice": true,
"supports_function_calling": true
}