All models
Text / ReasoningOpen weights8.2B

Vikasit 8B

Strong mid-range. Solid coding, analysis, and content generation.

Overview

Vikasit 8B is a strong mid-range generalist — coding, analysis, and content generation with reliable instruction following. The sweet spot for single-GPU deployment.

Specifications

Total parameters
8.2B
Architecture
Dense transformer
Layers
36
Attention
GQA (32 query / 8 KV heads)
Context window
32K native, 131K via YaRN
Vocabulary
151,669
Modalities
Text in → text out
License
Apache 2.0

Capabilities

  • Solid coding and code review
  • Document analysis and content generation
  • 131K extended context (YaRN)
  • Thinking and non-thinking modes
119 languages. Strong English + major Indian languages.

Benchmarks

BenchmarkScore
MMLU-Pro51.2
GPQA-Diamond62.0
AIME 202567.3
MATH-50097.4
LiveCodeBench v557.5
BFCL v368.1
IFEval85.0
HumanEvalN/A

Instruct numbers from the Qwen3 Technical Report; MMLU-Pro is the base-model figure. Thinking-mode scores shown.

Hardware & deployment

PrecisionMemory
bf16~16 GB
INT4~5 GB

Quick start

Vikasit 8B is an open-weight model. Self-host it with any OpenAI-compatible inference server and call it with the OpenAI SDK as shown below.

OpenAI-compatible Python (self-hosted, e.g. vLLM)
# pip install openai
import os
from openai import OpenAI

client = OpenAI(
    base_url="http://localhost:8000/v1",
    api_key="sk-local",  # self-hosted servers accept any token
)

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

print(resp.choices[0].message.content)

Limitations

  • Reasoning below 14B+ on hardest tasks
  • Single-GPU throughput limits high concurrency

Vikasit 8B FAQ

How much does Vikasit 8B cost?

Vikasit 8B is an open-weight model built on Qwen3-8B (Apache 2.0). Self-hosting the weights is free under the Apache 2.0 licence — you pay only for the hardware or cloud GPUs you run it on. Typical deployment fits the memory profiles listed in the hardware section above.

Is Vikasit 8B open weight?

Yes. Vikasit 8B is built on Qwen3-8B (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 run Vikasit 8B?

Because Vikasit 8B is open weight, you self-host it with any OpenAI-compatible inference server (such as vLLM or SGLang) loaded with the Qwen3-8B (Apache 2.0) weights, then call it with the OpenAI SDK by setting the base URL to your own endpoint.

What context window does Vikasit 8B support?

Vikasit 8B supports a 32K native, 131K via YaRN context window. It is a 8.2B Dense transformer model — full specifications are listed in the table above.

License & attribution

Apache 2.0

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