InternLM 2.5 20B.
Shanghai AI Lab's dense 20B open-weight. Strong long-context + tool use for its size.
1× Nvidia GeForce RTX 4080.
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 | |
|---|---|---|---|---|
| Self-hosted on rented GPU cluster | self hosted | — | — | Run yourself → |
| Self-hosted on rented GPU cluster | self hosted | — | — | Run yourself → |
| SiliconFlow | hosted inference | — | — | Launch ↗ |
| SiliconFlow | hosted inference | — | — | Launch ↗ |
Frequently asked.
How do I run InternLM 2.5 20B?
Where can I access InternLM 2.5 20B?
How much does it cost to run InternLM 2.5 20B?
Is InternLM 2.5 20B 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)
|
64 GB | — | — | Compare → | |
|
FP8
FP8 — 8-bit float (Hopper / Blackwell)
|
32 GB | $0.023/hr | — | Compare → | |
|
INT4
INT4 — 4-bit integer (~4× VRAM saving)
|
16 GB | — | — | Compare → |
What it costs per month across providers.
Estimate your monthly bill for InternLM 2.5 20B 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 InternLM 2.5 20B yet.
Rent the GPU instead of paying per token.
For an open-weights model like InternLM 2.5 20B, 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 InternLM 2.5 20B.
InternLM 2.5 20B is the dense flagship of Shanghai AI Lab's InternLM 2.5 series. 20B parameters, 1M-token context with the GROUP-INT extension, trained with a focus on long-form reasoning and agentic tool use. Apache-2.0 licence — fully open for commercial use. Often used as a strong open-weight option when Qwen and Llama don't fit the use case. Companion models include a 7B sibling and a math-specialist variant.