Environment & CO2 methodology

SuperCompress reduces KV cache prefill work by removing low-value tokens before inference. Fewer tokens means less GPU time, which means lower energy consumption and CO2 emissions for the same workflow.

These numbers are illustrative estimates, not per-deployment measurements. All assumptions are explicitly documented so you can adjust them for your hardware and energy mix.

What we measure

Metric	Definition
Tokens saved	original_tokens - kept_tokens per compression
KV savings %	(1 - kept/original) * 100
GPU-seconds avoided	Effective tokens saved / throughput
Wh saved	GPU-seconds * GPU watts / 3600
CO2 avoided	Wh * grid intensity (kg/kWh) / 1000
Water saved (est.)	Datacenter cooling water estimate (evaporative cooling)

Default assumptions

Defined in supercompress/benchmarks/metrics.py:

Parameter	Default	Rationale
tokens_per_gpu_second	2,500	7B-class model prefill on consumer GPU (e.g., RTX 4090)
gpu_watts	150 W	Typical single-GPU draw during inference (not peak)
kv_share_of_prefill	55%	Only the context/KV portion of prefill is attributed to savings
grid_kg_co2_per_kwh	0.417	US grid average (EIA 2024 data)

Python sustainability API

The sustainability_from_tokens_saved helper converts token savings into estimated energy and carbon metrics.

from supercompress.benchmarks.metrics import sustainability_from_tokens_saved

saved = result.original_tokens - result.kept_tokens
impact = sustainability_from_tokens_saved(saved)

print(impact.watt_hours_saved)
print(impact.co2_kg_avoided)
print(impact.water_liters_saved)
print(impact.assumptions.to_dict())

The impact.assumptions object exposes the four constants above so you can modify them for your specific hardware and region.

Scale examples

Based on our default assumptions, at ~800 tokens saved per compression:

Scale	Tokens avoided	kWh saved	CO2 avoided	Water saved (est.)
1 compression	~800	~0.00003	~0.00001 kg	~0.0001 L
1K compressions	~800K	~0.03	~0.01 kg	~0.1 L
1M compressions	~800M	~29	~12 kg	~100 L
10M compressions	~8B	~290	~120 kg	~1,000 L
100M compressions	~80B	~2,900	~1,200 kg	~10,000 L

Why CPU eviction matters

The learned policy (~5K parameters) runs on CPU before GPU inference. This means:

• The eviction itself consumes negligible energy — ~0.000001 kWh per compression on a modern CPU core
• But it avoids much larger GPU prefill work on long contexts
• At 65% token reduction, every 1 kWh spent on CPU eviction avoids ~1,000 kWh of GPU prefill

This 1:1000 energy ratio is why CPU-side eviction is environmentally significant. The tiny policy makes it possible — a 165M-parameter model running on CPU would consume much more energy per eviction, eroding the net savings.

Honesty for submissions

We follow these principles when presenting environmental impact data:

1. State assumptions clearly — never claim live datacenter metering unless you have it
2. Compare quality + savings together — truncation can save tokens but drops answers (25% oracle recall vs SuperCompress's 100%)
3. Don't claim CO2 without methodology — every estimate links back to the documented assumptions and formulas
4. CPU energy is negligible but included — the policy runs on CPU with ~60ms latency; the energy cost is orders of magnitude below the GPU prefill savings
5. Let users adjust constants — the Python API exposes all assumptions so users can plug in their own numbers

See benchmarks page for the interactive projection calculator and supercompress/benchmarks/metrics.py for the source code.