EmbeddingDense0.56B params514 max context

Multilingual E5 Large (1024-dim) VRAM Requirements: How Many GPUs & What It Costs

Multilingual E5 Large (1024-dim) needs 1.1 GB of VRAM for its weights at FP16 — 1.5 GB once a KV cache is loaded. Below is the full breakdown: which GPUs fit, how many you need, what it serves, and what it costs per million tokens. Every figure comes from tokcalc's open-source capacity model.

Weights (FP16)
1.1 GB
1.5 GB with KV cache
Fits on 1 GPU?
Yes
RTX 3080 10GB or larger
Cheapest config
$0.0105
1× RTX 3080 10GB / 1M tokens
Fastest config
139286
tok/s on 1× Cerebras CS-3 (WSE-3)
VRAM requirement by quantization
Weight footprint = params × bytes per param. The “with KV cache” column adds one request at 4K context.
QuantizationWeightsWith KV cacheSmallest single GPU
FP16recommended1.1 GB1.5 GBRTX 3080 10GB (10 GB)
FP8 (E4M3)0.6 GB1.0 GBRTX 3080 10GB (10 GB)
AWQ 4-bit0.3 GB0.7 GBRTX 3080 10GB (10 GB)
Which GPU should you use?
Reference workload: 8K context, 8 concurrent users, continuous batching, FP16. Ranked by fitting first, then by tokens per dollar-hour.
GPUGPUsTotal VRAMDecode tok/sAggregate tok/sCost / 1M tokens
RTX 3080 10GB1×10 GB441.15293$0.0105
RTX 3090 24GB1×24 GB543.26519$0.0107
RTX 5090 32GB1×32 GB104012480$0.0111
RTX 4090 24GB1×24 GB585.07020$0.0158
RTX 4080 16GB1×16 GB416.14993$0.0167
AMD MI300X 192GB1×192 GB307636911$0.0188
V100 SXM2 32GB1×32 GB522.36268$0.0222
A100 40GB SXM41×40 GB899.610795$0.0232
AMD MI325X 288GB1×288 GB348241786$0.0233
A100 80GB SXM41×80 GB116113929$0.0239
A10G 24GB1×24 GB348.24179$0.0239
Intel Gaudi 3 128GB1×128 GB191522982$0.0242
How concurrency changes with context length
On 1× RTX 3080 10GB at FP16. KV cache scales linearly with context, so long contexts cut the number of users you can serve from the same GPUs.
ContextConcurrent usersKV per requestFits?
4K22412.3 MB✓ fits
8K11824.6 MB✓ fits
33K23.22 GB✓ fits
66K16.44 GB✓ fits
131K012.88 GB✗ does not fit
Frequently asked questions

Can Multilingual E5 Large (1024-dim) run on a single GPU?

Yes. At FP16 the model needs 1.5 GB of VRAM, so it fits on a single RTX 3080 10GB or anything with more VRAM. At full FP16 it needs 1.5 GB, which fits on one RTX 3080 10GB or larger.

How many GPUs does Multilingual E5 Large (1024-dim) need?

At FP16 with 8,192-token context and 8 concurrent users, the cheapest configuration that fits is 1× RTX 3080 10GB (10 GB total VRAM). $0.0105 per 1M output tokens.

How fast can Multilingual E5 Large (1024-dim) be served?

On 1× Cerebras CS-3 (WSE-3) at FP16, the aggregate throughput is 139286 tokens/sec with continuous batching. That hardware has no published rental price, so its cost per million tokens is not comparable.

How much VRAM does Multilingual E5 Large (1024-dim) need?

Weights alone: 1.1 GB at FP16, 1.1 GB at FP16. Add KV cache for your context length and concurrency — every concurrent request at 8,192 tokens costs extra VRAM, so budget headroom above the weight figure.

Is it cheaper to self-host Multilingual E5 Large (1024-dim) than to call an API?

At 1× RTX 3080 10GB the marginal cost is $0.0105 per 1M output tokens, before you add idle capacity, ops staff, or egress. That is typically cheaper than frontier API pricing at sustained volume, but API calls win when your load is bursty or too small to keep a GPU busy. The Build vs Buy tab on tokcalc prices this against a specific provider for you.

Try Multilingual E5 Large (1024-dim) with your own context and concurrency

The calculator uses the exact same formulas as this page — change the workload and watch every number update.

Open in calculator

Multilingual E5 Large (1024-dim) on specific GPUs

Multilingual E5 Large (1024-dim) on RTX 3080 10GBMultilingual E5 Large (1024-dim) on RTX 3090 24GBMultilingual E5 Large (1024-dim) on RTX 5090 32GBMultilingual E5 Large (1024-dim) on RTX 4090 24GBMultilingual E5 Large (1024-dim) on RTX 4080 16GBMultilingual E5 Large (1024-dim) on AMD MI300X 192GBMultilingual E5 Large (1024-dim) on V100 SXM2 32GBMultilingual E5 Large (1024-dim) on A100 40GB SXM4
Model architecture
The numbers behind the KV-cache and throughput math.
Total parameters0.56B
Active parameters per token0.56B
ArchitectureDense
Attention layoutFull attention (GQA)
Transformer layers24
Hidden dimension1024
Attention heads16 query / 16 KV (GQA)
Head dimension64
Vocabulary size250,047
Max context (catalog)514
KV cache per token @ 8K96.0 KB

Every layer caches keys and values for the whole context, so the KV cache grows linearly with conversation length.