Yi-34B.
Earlier open-weight Yi release — bilingual EN/ZH, 32K-token context.
1× Nvidia RTX A5000.
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 | |
|---|---|---|---|---|
| 01.AI Platform | api direct | — | — | Launch ↗ |
| Self-hosted on rented GPU cluster | self hosted | — | — | Run yourself → |
| Self-hosted on rented GPU cluster | self hosted | — | — | Run yourself → |
| Together AI | hosted inference | — | — | Launch ↗ |
| Together AI | hosted inference | — | — | Launch ↗ |
Variants in the Yi family.
Frequently asked.
How do I run Yi-34B?
Where can I access Yi-34B?
How much does it cost to run Yi-34B?
Is Yi-34B 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)
|
94 GB | — | — | Compare → | |
|
FP8
FP8 — 8-bit float (Hopper / Blackwell)
|
48 GB | $0.28/hr | — | Compare → | |
|
INT4
INT4 — 4-bit integer (~4× VRAM saving)
|
24 GB | — | — | Compare → |
What it costs per month across providers.
Estimate your monthly bill for Yi-34B across every host that publishes per-token pricing. Slide your token volumes; the chart + table re-rank cheapest-first.
No priced API access rows on file for Yi-34B yet.
Rent the GPU instead of paying per token.
For an open-weights model like Yi-34B, 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.
About Yi-34B.
Yi-34B is 01.AI's mid-size open-weight model. 34B parameters, dense transformer, trained on a bilingual EN/ZH corpus. Apache-2.0 licence makes it permissive enough for commercial fine-tuning. Released alongside a 6B sibling and a 200K-context variant (Yi-6B-200K) widely used for long-document workloads. Largely superseded by closed Yi-Lightning for production but still a popular fine-tuning base.