impulse

Vision Small

API id vision-small
Input / 1M
$1.05
at Vispark
Output / 1M
$3.16
at Vispark
Context window
1M
tokens
Max output
66K
tokens per reply
Hosts
1
1 listings
Released
-
date not published

Pricing

List prices at Vispark, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$1.05$3.16
Cached input
100% off input
$0.00

What it costs a month

At Vispark list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$458
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$366
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$568
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$3,278
Use your own numbers

Where to run it

1 host serve Vision Small. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
VisparkCloud
vision-small
$1.05 / $3.16

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
AudioImagesTextVideoPDFs
Produces
Text
Features
Parallel tool callsPDF input

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 Vispark.

Smaller Independent models

No cheaper Independent model keeps its key capabilities.

Other frontier labs

  • Gemini 3.8 Flash
    Google · on Google AI Studio
    $0.75 / $3.75
    Has every key capability
    5% less
  • Grok Build 0.1
    xAI · on xAI
    $1.00 / $2.00
    Has every key capability
    21% less
  • GPT-5.4 Nano
    OpenAI · on OpenAI
    $0.20 / $1.25
    Has every key capability
    71% less
  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    87% less

Open-weight picks

  • Kimi K2.7 Code
    Moonshot AI · on CLF AI Gateway
    $0.38 / $1.60
    Has every key capability
    57% less
  • Inkling Small
    Independent · on DeepInfra
    $0.45 / $1.20
    Has every key capability
    60% less
  • DeepSeek Flash
    DeepSeek · on DeepSeek
    $0.30 / $1.20
    Has every key capability
    67% less
  • Qwen3.7 Plus
    Alibaba Qwen · on Replicate
    $0.276 / $1.10
    Has every key capability
    69% less

Call it

Using Vispark's API.

Python
from litellm import completion

response = completion(
    model="vispark/vision-small",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Vision Small cost?

On Vispark, Vision Small costs $1.05 per million input tokens and $3.16 per million output tokens.

Which APIs offer Vision Small?

Vision Small runs on 1 API: Vispark.

What is Vision Small's context window?

1M tokens, with up to 66K tokens in one reply.

Is Vision Small being retired?

No retirement date is scheduled for it on any API we track.

What are cheaper alternatives to Vision Small?

Models that cost less and keep most of what it can do: Gemini 3.8 Flash on Google AI Studio (5% less), Grok Build 0.1 on xAI (21% less) and Kimi K2.7 Code on CLF AI Gateway (57% less). Savings are on a mix of 3 input tokens to 1 output token.

What is Vision Small's API model id?

Use vision-small with Vispark.

Sources

Raw catalog entry for vision-small on Vispark
{
  "cache_read_input_token_cost": 0,
  "input_cost_per_token": 0.00000105,
  "litellm_provider": "vispark",
  "max_input_tokens": 1000000,
  "max_output_tokens": 65536,
  "mode": "chat",
  "output_cost_per_token": 0.00000316,
  "source": "https://api.lab.vispark.in/v1/models",
  "supported_endpoints": [
    "/v1/chat/completions",
    "/v1/completions"
  ],
  "supported_modalities": [
    "text",
    "image",
    "audio",
    "video"
  ],
  "supported_output_modalities": [
    "text"
  ],
  "supports_function_calling": true,
  "supports_native_streaming": true,
  "supports_parallel_function_calling": true,
  "supports_pdf_input": true,
  "supports_reasoning": true,
  "supports_response_schema": true,
  "supports_system_messages": true,
  "supports_tool_choice": true,
  "supports_vision": true
}