How much memory does sarvam-105b need?

About 61 GB at 4-bit, 114 GB at 8-bit and 214 GB at full 16-bit precision, including an 8,192-token KV cache and about 1.5 GB of runtime overhead. These are estimates from its parameter count.

Can sarvam-105b run on a 24 GB graphics card?

Not entirely: even at 4-bit it needs about 61 GB, so part of it would sit in system memory, which is much slower.

What is sarvam-105b’s context window?

Its configuration sets 131,072 tokens. Longer contexts need more memory for the KV cache, so the practical limit on a given machine can be lower.

ANVE: the private AI office, on the record. See what that means

ANVE

Run it on your own hardware · updated 2026-10-16

sarvam-105b
The memory it needs, what you may do with it, and the hardware it suits.

Short answer. sarvam-105b needs about 61 GB of memory at 4-bit, 114 GB at 8-bit and 214 GB at full 16-bit precision, including runtime overhead and an 8,192-token KV cache. It has about 106 billion parameters and a 131,072-token context window. Licence in short: commercial use allowed. By memory alone, on ANVE’s launch configurations: ANVE One does not fit, ANVE Office only with partial offload to system memory, which is much slower, ANVE Office Pro (GPU) only with partial offload to system memory, which is much slower. These are estimates from its published size, not benchmarks.

What it is

sarvam-105b is a language model for chat, drafting and question answering, published by Sarvam AI on Hugging Face on 3 March 2026. It has about 106 billion parameters and a configured context window of 131,072 tokens. It is a mixture-of-experts model: 128 experts, 8 used for each token.

Its card lists the licence as Apache 2.0. In short, the licence allows commercial use, modification and redistribution, with attribution and a copy of the licence, and includes a patent grant.

Hugging Face counted about 24,722 downloads of it in the 30 days before we checked (7 Oct 2026), and 309 people have liked it there.

The card lists 23 languages: en, hi, bn, ta, te, mr, gu, kn, ml, pa, or, as and 11 more (ISO codes as given on the card).

How much memory it needs

How these figures are worked out: weights = parameters × bits per weight ÷ 8. The KV cache is added for 8,192 tokens of context, plus about 1.5 GB of runtime overhead. Treat the totals as planning estimates.

Each token of context adds about 0.04 MB of KV cache at 16-bit, so 32,768 tokens would add about 1.2 GB. This model uses multi-head latent attention, which stores a compressed cache.

Precision	Weights	KV cache (8,192 tokens)	Total, estimated
BF16 (full precision)	212 GB	0.30 GB	214 GB
8-bit (Q8_0)	113 GB	0.30 GB	114 GB
4-bit (Q4_K)	60 GB	0.30 GB	61 GB

Architecture at a glance

Setting	Value in config.json
Architecture class	SarvamMLAForCausalLM
Model type	sarvam_mla
Layers	32
Hidden size	4,096
Attention heads	64
Head size	576
Vocabulary	262,144
Experts	128, 8 active per token
Configured context	131,072
Stored precision	float32

Will it fit on ANVE?

Arithmetic against the launch specification, not a test: about three quarters of ANVE One’s 64 GB of unified memory for the model (as AMD’s Variable Graphics Memory allows), and the 24 GB on the graphics card in ANVE Office and ANVE Office Pro, which keep the rest of the stack in 64 GB and 128 GB of system memory. ANVE’s tested default is Qwen 3.8 27B; this model is not yet validated on ANVE.

Configuration (at launch)	Memory for the model	Verdict
ANVE One	about 48 GB of 64 GB unified memory	does not fit
ANVE Office	24 GB on the RTX PRO 4000 Blackwell SFF Edition	only with partial offload to system memory, which is much slower
ANVE Office Pro (GPU)	24 GB on the RTX PRO 4000 Blackwell	only with partial offload to system memory, which is much slower

What an office would use it for

Models this large are stronger on long, hard drafting and reasoning, but need most of a 64 to 128 GB machine even at 4-bit, which leaves less room for long documents and several users.

Similar models to compare

c4ai-command-r-plus (103.8B parameters, about 60 GB at 4-bit before context; non-commercial only)
GLM-4.5-Air (110.5B parameters, about 64 GB at 4-bit before context; commercial use allowed)
gpt-oss-120b (116.8B parameters, about 67 GB at 4-bit before context; commercial use allowed)

Questions people ask

How much memory does sarvam-105b need?: About 61 GB at 4-bit, 114 GB at 8-bit and 214 GB at full 16-bit precision, including an 8,192-token KV cache and about 1.5 GB of runtime overhead. These are estimates from its parameter count.
Can sarvam-105b run on a 24 GB graphics card?: Not entirely: even at 4-bit it needs about 61 GB, so part of it would sit in system memory, which is much slower.
Is sarvam-105b free for commercial use?: Its card lists Apache 2.0, which allows commercial use, modification and redistribution, with attribution and a copy of the licence, and includes a patent grant. Read the licence itself before relying on this summary; it is not legal advice.
What is sarvam-105b’s context window?: Its configuration sets 131,072 tokens. Longer contexts need more memory for the KV cache, so the practical limit on a given machine can be lower.

What is not real yet

No sealed ANVE unit exists; every image of the appliance on this site is a labelled render. The software runs today on a development machine. No customers yet, no certifications, no delivery date, and the final price is not announced.

Reservations are ₹999 in India or US$12 elsewhere, refundable in full until you confirm an order in writing (terms). Offices that cannot use cloud AI can apply to be one of ten design partners.

sarvam-105bThe memory it needs, what you may do with it, and the hardware it suits.