Swe 1.7
swe-1.7Pricing
List prices at Cognition, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $0.50 | $2.50 | |
| Cached input 60% off input | $0.20 | 60% off input |
What it costs a month
At Cognition list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$311
- RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$258
- Coding agent1,000 requests a day, 40K in / 2K out, cached$498
- Overnight batch20,000 requests a day, 4K in / 0.4K out$1,800
Where to run it
1 host serve Swe 1.7. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
CognitionCloud swe-1.7 | $0.50 / $2.50 | swe-1.7 |
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
- Features
- Prompt caching
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 Cognition.
Smaller Independent models
No cheaper Independent model keeps its key capabilities.
Other frontier labs
- 15% less
- 54% less
- 80% less
- 80% less
Open-weight picks
- 13% less
- 14% less
- 21% less
- 31% less
Model line
Newer and older versions, at each lab's own price.
- UndatedSwe 1.7Latest$0.50 / $2.50
- UndatedSwe 1.6$0.50 / $2.50
Call it
Using Cognition's API.
from litellm import completion
response = completion(
model="cognition/swe-1.7",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does Swe 1.7 cost?
On Cognition, Swe 1.7 costs $0.50 per million input tokens and $2.50 per million output tokens. Cached input costs $0.20 per million tokens.
Which APIs offer Swe 1.7?
Swe 1.7 runs on 1 API: Cognition.
Is Swe 1.7 being retired?
No retirement date is scheduled for it on any API we track.
Is there a newer version of Swe 1.7?
No. Swe 1.7 is the newest model in its line.
What are cheaper alternatives to Swe 1.7?
Models that cost less and keep most of what it can do: Gemini 3.5 Flash Lite on Google AI Studio (15% less), GPT-5.4 Nano on OpenAI (54% less) and Motif 3 on ai& (13% less). Savings are on a mix of 3 input tokens to 1 output token.
What is Swe 1.7's API model id?
Use swe-1.7 with Cognition.
Sources
Raw catalog entry for swe-1.7 on Cognition
{
"input_cost_per_token": 5e-7,
"output_cost_per_token": 0.0000025,
"cache_read_input_token_cost": 2e-7,
"litellm_provider": "cognition",
"mode": "chat",
"supports_function_calling": true,
"supports_prompt_caching": true,
"source": "https://docs.devin.ai/desktop/models"
}