All models
Vision-LanguageAvailable via API8B

Vikasit Vision 8B

Strong vision. Complex image reasoning, visual code generation.

Overview

Vikasit Vision 8B is the strongest vision-language model in the lineup — complex image reasoning, document intelligence, and visual code generation. Served live via the Vikasit API.

Specifications

Total parameters
8B
Architecture
Dense ViT + LLM, interleaved-MRoPE, DeepStack multi-level features
Context window
256K native, ~1M expandable
Modalities
Text + image + video in → text out
License
Apache 2.0

Capabilities

  • Complex visual reasoning
  • Document intelligence and OCR (32 languages)
  • Visual code generation (UI → code)
  • Long video understanding
OCR across 32 languages. (No Indic-script OCR claimed by the base model.)

Benchmarks

BenchmarkScore
MMMU (val)74.1
MMMU-Pro60.4
DocVQA (test)95.3
ChartQA (test)88.6
MathVista (mini)81.4
AI2D84.9
OCRBench819
RealWorldQA73.5
Video-MME71.8
TextVQAN/A

Numbers from the Qwen3-VL Technical Report (arXiv:2511.21631, Table 4), thinking-mode column.

Hardware & deployment

PrecisionMemory
bf16~18 GB
INT4~6 GB

Quick start

Call Vikasit Vision 8B through the OpenAI-compatible Vikasit AI API at https://api.vikasit.ai/v1 using the model id vikasit-3-vision.

OpenAI-compatible Python (Vikasit AI API)
# pip install openai
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.vikasit.ai/v1",
    api_key=os.environ["VIKASIT_API_KEY"],
)

resp = client.chat.completions.create(
    model="vikasit-3-vision",
    messages=[
        {"role": "user", "content": "Explain Vikasit Vision 8B in one sentence."}
    ],
)

print(resp.choices[0].message.content)
# or with curl
curl https://api.vikasit.ai/v1/chat/completions \
  -H "Authorization: Bearer $VIKASIT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "vikasit-3-vision",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Limitations

  • No Indic-script OCR (base supports 32 listed languages)

Vikasit Vision 8B FAQ

How much does Vikasit Vision 8B cost?

Vikasit Vision 8B is served through the Vikasit AI API on usage-based, pay-as-you-go pricing billed per million input and output tokens — see the Vikasit AI pricing page for current rates. Because it is built on the open-weight Qwen3-VL-8B (Apache 2.0), you can also self-host the weights for free under the Apache 2.0 licence and pay only for your own compute.

Is Vikasit Vision 8B open weight?

Yes. Vikasit Vision 8B is built on Qwen3-VL-8B (Apache 2.0) and distributed under the Apache 2.0 licence, so the weights are openly available for self-hosting, fine-tuning, and commercial use, subject to the upstream licence terms.

How do I use Vikasit Vision 8B with the OpenAI SDK?

The Vikasit AI API is OpenAI-compatible. Point the OpenAI client's base URL at https://api.vikasit.ai/v1, set your Vikasit API key, and pass "vikasit-3-vision" as the model. The quick-start snippet above shows the exact Python call.

What context window does Vikasit Vision 8B support?

Vikasit Vision 8B supports a 256K native, ~1M expandable context window. It is a 8B Dense ViT + LLM, interleaved-MRoPE, DeepStack multi-level features model — full specifications are listed in the table above.

License & attribution

Apache 2.0

Built on Qwen3-VL-8B (Apache 2.0). Upstream copyright, license, and attribution notices are retained.

Qwen3-VL Technical Report (arXiv:2511.21631)