Sarvam 30B
Efficient Indian-language reasoning model for chat, coding, and multilingual work
Sarvam 30B
Efficient Indian-language reasoning model for chat, coding, and multilingual work
- • FP16 Weights = 30.0B × 2B = 60.00 GB
- • INT4 Weights = 30.0B × 0.55B = 16.50 GB
- • KV Cache (128000 ctx, FP16) ≈ 24.41 GB
- • Activation Buffer = ~20% overhead
Hardware & Execution ParametersCode
Genealogical Graph & Evolutionary Provenance
Tracing foundational base architecture ancestry, architectural successors, scale siblings, and reasoning distillation derivatives.
Sarvam 30B operates as an autonomous foundation model architecture without direct precursor derivatives in this catalog.
Architecture Engineering & Capability Deep-Dive
An objective architectural evaluation of Sarvam 30B by Sarvam, analyzing underlying compute dynamics, memory constraints, and deployment economics.
Topology & Attention Mechanics
A robust autoregressive transformer utilizing standard attention patterns for predictable and coherent token generation.
Evaluation Profile & Reasoning
Exhibits frontier-tier behavior in reasoning and coding.
LLM Hardware Sizing & Serving
For open deployments via vLLM/SGLang, quantization (INT4/AWQ) is heavily recommended to fit dense memory constraints, or multi-GPU pipeline parallelism for full FP16.
Inference Economics & Workflows
Well-suited for enterprise pipelines where capability is balanced against per-million token costs.
Architectural Strengths vs. Considerations
An objective balance sheet analyzing the operational advantages and production constraints of deploying Sarvam 30B.
Key Architectural Strengths
- Massive 128,000-token context allows full-repository and book-length ingestion.
Operational Considerations
- 128k+ token prefill stages become heavily compute-bound and balloon KV cache without PagedAttention chunking.
Inference Runtimes & Hardware Sizing
Deployment targets, inference engines, and memory requirements for Sarvam 30B.
python3 -m vllm.entrypoints.openai.api_server --model Sarvam 30B --tensor-parallel-size 1 --gpu-memory-utilization 0.95 --max-model-len 4096 --enable-chunked-prefillollama run sarvam 30bpython3 -m sglang.launch_server --model-path Sarvam 30B --tp 1 --trust-remote-codetext-generation-launcher --model-id Sarvam 30B --num-shard 1 --max-batch-prefill-tokens 32000High-throughput PagedAttention server
One-click CLI & local desktop serving
Fast multi-turn structured decoding
Text Generation Inference
GGUF CPU/Apple Silicon execution
~72.0 GB VRAM required
~36.0 GB VRAM (Hopper speedup)
~19.8 GB VRAM
CPU RAM / Apple Silicon optimized
API & Deployment Pricing
Open-weights model available for local and private cloud deployment. Compute costs depend on the target GPU hardware instance.
| Deployment Tier | Pricing Structure |
|---|---|
| Open Checkpoint Weights | $0.00 (Free Download) |
| Inference Token Consumption | $0.00 / Token |
Comparable Foundation Architectures
Alternative models in the Code class with similar capabilities, context windows, or deployment profiles.
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ.get("SARVAM_KEY", "EMPTY"),
base_url="http://localhost:8000/v1"
)
response = client.chat.completions.create(
model="sarvam-sarvam-30b",
messages=[{"role": "user", "content": "Explain quantum superposition in 2 sentences."}]
)
print(response.choices[0].message.content)Frequently Asked Questions about Sarvam 30B
Essential facts, architectural specs, hardware constraints, and pricing answers for Sarvam 30B.
To run Sarvam 30B (30B) locally, you generally need Depends on quantization. We recommend using quantized GGUF/AWQ formats with Ollama or vLLM to optimize memory footprint.
All technical specifications, parameter distributions, context architectures, and benchmark evaluations for Sarvam 30B are audited against primary source release documentation, research whitepapers, and verified vendor API endpoints.