Command R 35B KV-Cache Calculator
Per-token and total key-value cache memory for Command R 35B across context length, batch size and cache precision.
Command R 35B: ~160.0 KB per token at FP16. Cohere's Command R 35B is tuned for RAG and tool use at 128K context; with 8 KV heads on a wide 8192 hidden size, cache planning dominates deployment decisions.
Formula
Disclaimer: This tool is for general informational and estimation purposes only and is not professional financial, tax, accounting or legal advice. All figures are estimates — verify with a qualified professional before making decisions. Read the full disclaimer.
About Command R 35B KV-Cache Calculator
The KV cache is the hidden memory cost of serving Command R 35B: every generated or prompted token stores its attention keys and values for reuse, and at long contexts this cache can rival the model weights themselves. This calculator uses Command R 35B's exact attention geometry (8 KV heads × 128-dim heads × 40 layers) to give per-token, per-sequence and whole-batch cache sizes at FP16, FP8 and INT4 precision. Use it to size batch limits for your GPU or to see what a 128K-context request really costs.
How to use Command R 35B KV-Cache Calculator
- 1Enter your values into Command R 35B KV-Cache Calculator — sensible, domain-typical defaults are pre-filled so you see a real result immediately.
- 2The result recomputes live using the formula shown on the page; there is no button to press.
- 3Adjust any input to compare scenarios, then read the worked example to see the substituted numbers.
Why use Command R 35B KV-Cache Calculator?
- ✓Computes Command R 35B KV-Cache instantly in your browser — no sign-up, no upload, no server round-trip.
- ✓100% free and unlimited, with the exact formula shown: cache/token = 2(K,V) × layers × kv_heads × head_dim × bytes = 2 × 40 × 8 × 128 × bytes.
- ✓Runs entirely client-side, so every value you enter stays private on your device.
- ✓Live recompute as you type, with a worked example and authoritative references for trust.
Frequently asked questions
What makes Command R's memory profile RAG-friendly?+
RAG stuffs long retrieved passages into the prompt, so cache cost per token matters more than weight size. With GQA (8 KV heads × 128 head_dim), a 64K-token RAG context costs ~5.2 GB at FP16 — manageable next to 70 GB of FP16 weights.
Single-GPU options for Command R 35B?+
FP16 weights (~70 GB) need an 80 GB card; INT8 (~35 GB) fits a 48 GB A6000/L40S with cache headroom; 4-bit (~18 GB) runs on a 24 GB card for short-to-medium contexts. Long-context RAG pushes you to 48 GB+.
Why does the KV cache matter more than weights for serving throughput?+
Weights are paid once per GPU; cache is paid per concurrent request and per token of context. Batch size — and therefore throughput — is capped by how many sequence caches fit in the VRAM left after weights, which is exactly what this tool computes.
What does paged attention change?+
PagedAttention (vLLM) allocates the cache in fixed-size blocks on demand instead of reserving the full context up front, eliminating fragmentation and letting you overcommit. The per-token cost shown here is unchanged — you just stop paying for unused reservation.
Related ML & AI tools
ROC-AUC Calculator (from TPR/FPR points)
Trapezoidal area under the ROC curve from your (FPR, TPR) operating points — the threshold-independent ranking score.
● LiveClassification Threshold Cost Calculator
Find the probability cutoff that minimizes expected cost given your false-positive and false-negative penalties.
● LiveSilhouette Score Calculator
Cluster cohesion vs separation for one point — the building block of the silhouette metric for choosing K.
● Live