Llama 3.1 8B.
Meta's most popular open-weight small LLM — fits anywhere.
1× Nvidia P102-100.
Most-aggressive quantisation we have a working recommendation for. Lower precision = less VRAM = cheaper hardware, at a small accuracy cost.
Cheapest hosted endpoints.
| Provider | Access | $/M in | $/M out | |
|---|---|---|---|---|
| OpenRouter | api aggregator | $0.02 | $0.05 | Launch ↗ |
| Groq | hosted inference | $0.05 | $0.08 | |
| Together AI | hosted inference | $0.18 | $0.18 | |
| Fireworks AI | hosted inference | $0.2 | $0.2 |
What it's best at.
Speed across providers.
Tokens/sec and time-to-first-token measured against the same prompt template on each provider's API.
| Provider | Tokens/sec | TTFT | Total |
|---|---|---|---|
| OpenRouter | 50.0 | 738 ms | 2559 ms |
Variants in the Llama family.
Llama 3.3 70B
Meta's best-in-class open-weight LLM — 70B class.
Llama 3.1 70B
Llama 3.1 70B — production workhorse, superseded by 3.3 but still widely depl...
Llama 3.1 405B
Meta's largest open-weight LLM — dense 405B, frontier-class at launch.
Llama 3.2 90B Vision
Meta's largest vision-capable Llama.
Llama 3.2 11B Vision
Meta's open-weight multimodal LLM — vision + text in 11B.
Llama 3.2 3B
3B Llama — laptop-class chat + RAG.
Llama 3.2 1B
Meta's smallest Llama — mobile + on-device target.
Frequently asked.
How do I run Llama 3.1 8B?
Where can I access Llama 3.1 8B?
How much does it cost to run Llama 3.1 8B?
Is Llama 3.1 8B open-source or proprietary?
Cheapest hardware per quantisation.
Each row is one quantisation tier (the same weights compressed differently). Lower precision → lower VRAM → cheaper hardware, at the cost of small accuracy loss. $/hr refreshed hourly from each provider's API.
| Quantisation | Cheapest GPU config | Total VRAM | Live $/hr | tokens/sec | |
|---|---|---|---|---|---|
|
FP16
FP16 — half precision (default)
|
Comfortable single-GPU local target.
|
24 GB | — | — | Compare → |
|
FP8
FP8 — 8-bit float (Hopper / Blackwell)
|
10 GB | — | — | Compare → | |
|
INT8
INT8 — 8-bit integer
|
Older T4s on a budget.
|
16 GB | $0.070/hr | — | Compare → |
|
INT4
INT4 — 4-bit integer (~4× VRAM saving)
|
5 GB | — | — | Compare → |
What it costs per month across providers.
Estimate your monthly bill for Llama 3.1 8B across every host that publishes per-token pricing. Slide your token volumes; the chart + table re-rank cheapest-first.
Cheapest provider on the left.
Total monthly cost — input + output tokens combined.
Bill breakdown.
| Provider | Monthly total | |
|---|---|---|
| $0.3 | Sign up ↗ | |
| $0.66 | ||
| $2.16 | ||
| $2.4 |
Rent the GPU instead of paying per token.
For an open-weights model like Llama 3.1 8B, you can rent a GPU and serve inference yourself. The math: cheapest GPU rental × 730 hours/month + your electricity rate × power draw.
Assumes the GPU runs 24/7 at ~85% utilisation. If your traffic is bursty, you'll pay less for the API and probably more for the GPU (idle hours still cost rental). The breakeven analysis lives on the Self-host vs API breakeven tool.
What it's best at.
Scores normalised against benchmark ceilings (100 = perfect). Coloured by tier — coral 80+ frontier, lavender 65+ strong, sage 50+ solid, slate below.
Published scores.
| Benchmark | Score | Source |
|---|---|---|
| GPQA | 32.8 | official ↗ |
| MMLU | 73.0 | official ↗ |
| HumanEval | 72.6 | official ↗ |
About Llama 3.1 8B.
Llama 3.1 8B is by far the most-downloaded open-weight model on Hugging Face. 8B parameters, 128K context, MIT-adjacent Community License, runs on any GPU with ≥16 GB VRAM at FP16 or ≥6 GB at INT4. The reference 8B model — every other 7-9B open-weight (Mistral 7B, Qwen 2.5 7B, Phi-3.5) is compared against it.