meta-llama · text

Llama-2-7b-chat-hf

meta-llama/Llama-2-7b-chat-hf

Llama-2-7b-chat-hf at Q4_K_M is exactly 4,081,004,224 bytes (3.80 GiB / 4.08 GB) — an effective 4.845 bits per weight, not the nominal 4. Its KV cache at 32K is 16.00 GiB.

From the file· summed from 1 file(s)From the file· KV from mirror (mirror:NousResearch/Llama-2-7b-chat-hf)
Parameters
6.7B
Architecture
llama
32 layers
Context
4,096
native (config.json)
License
llama2

Shipped quantizations

● exact bytes, summed from published files
QuantSize●Exact bytes●Effective bpw●Tensors●Publisher
Q2_K2.63 GiB2,825,940,6723.355—TheBloke
Q2_K2.63 GiB2,825,941,6003.355—second-state
Q3_K_S2.75 GiB2,948,304,5763.500—TheBloke
Q3_K_S2.75 GiB2,948,305,5043.500—second-state
Q3_K_M3.07 GiB3,298,004,6723.916291TheBloke
Q3_K_M3.07 GiB3,298,005,6003.916—second-state
Q3_K_L3.35 GiB3,597,110,9764.271—TheBloke
Q3_K_L3.35 GiB3,597,111,9044.271—second-state
Q4_03.56 GiB3,825,807,0404.542291TheBloke
Q4_03.56 GiB3,825,807,9684.542—second-state
Q4_K_S3.59 GiB3,856,740,0324.579—TheBloke
Q4_K_S3.59 GiB3,856,740,9604.579—second-state
Q4_K_M3.80 GiB4,081,004,2244.845291TheBloke
Q4_K_M3.80 GiB4,081,005,1524.845—second-state
Q5_04.33 GiB4,651,691,7125.523—TheBloke
Q5_K_S4.33 GiB4,651,691,7125.523—TheBloke
Q5_K_S4.33 GiB4,651,692,6405.523—second-state
Q5_04.33 GiB4,651,692,6405.523—second-state
Q5_K_M4.45 GiB4,783,156,9285.679291TheBloke
Q6_K5.15 GiB5,529,194,1766.564291TheBloke
Q6_K5.15 GiB5,529,195,1046.564—second-state
Q8_06.67 GiB7,161,089,7288.502291TheBloke
Q8_06.67 GiB7,161,090,6568.502—second-state
Q5_K_M2 shards8.91 GiB9,566,314,78411.357—second-state
F1612.55 GiB13,478,105,69616.002—second-state

KV cache by context

computed per layer
ContextKV cache (f16)●Flat formulaOverstated byFull / windowed / recurrent
4,0962.00 GiB2.00 GiB—32 / 0 / 0
8,1924.00 GiB4.00 GiB—32 / 0 / 0
16,3848.00 GiB8.00 GiB—32 / 0 / 0
32,76816.00 GiB16.00 GiB—32 / 0 / 0
65,53632.00 GiB32.00 GiB—32 / 0 / 0
131,07264.00 GiB64.00 GiB—32 / 0 / 0

Compare with

same modality, comparable size
vs gemma-4-12B-it-qat-q4_0-unquantizedvs gemma-4-E4B-itvs Qwen3-4Bvs Qwen3-8Bvs HyperCLOVAX-SEED-Text-Instruct-1.5Bvs Llama-3.2-1B-InstructAll alternatives →Will it fit on my card? →

Will it run on your card?

full quant x context sweep
By VRAM:8GB12GB16GB24GB32GB48GB
Radeon RX 6500 XT 4GBGeForce RTX 3050 6GBGeForce RTX 5050 8GBGeForce RTX 3080 10GBGeForce RTX 2080 Ti 11GBGeForce RTX 5070 12GBGeForce RTX 5060 Ti 16GBApple M3 Pro 18GBGeForce RTX 3080 Ti 20GBGeForce RTX 5090 D V2 24GBGeForce RTX 5090 D 32GBApple M5 Max 36GBApple M5 Pro 48GBApple M5 Max 64GBApple M3 Ultra 96GBApple M5 Max 128GBApple M2 Ultra 192GBApple M3 Ultra 256GBApple M3 Ultra 512GB

Why other calculators give a different number

A parameters × bits ÷ 8 estimate puts Q4_K_M at roughly 3.53 GiB. The real file is 3.80 GiB, because a quantization is a mixture and some tensors are always kept at higher precision.

Architecture

from mirror:NousResearch/Llama-2-7b-chat-hf
Layers
32
Attention heads
32
KV heads
32
Head dim
128
Hidden size
4096
Vocab
32,000
Sliding window
none
SWA period
—
MLA
no
Experts
—
Experts per token
—
use_sliding_window
—

Questions people ask

How much VRAM does Llama-2-7b-chat-hf need?
Q4_K_M is exactly 4,081,004,224 bytes (3.80 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
How large is Llama-2-7b-chat-hf's KV cache?
16.00 GiB at 32K context with an f16 cache, computed per layer. Quantizing the cache to q8_0 roughly halves it, which is often the difference between a context length fitting and not.
Which quantization of Llama-2-7b-chat-hf should I use?
Q4_K_M is the usual default. Pick the largest quantization that fits your card at the context you actually need — the table above gives exact sizes for every one published.
Llama-2-7b-chat-hf — VRAM requirements, exact quant sizes — ossmodeldb