Mistral Nemo

Mistral-Nemo-Instruct-2407

Mistral AI · open-permissive apache-2.0

Parameters (B)12.248
Architecturedense
Layers / KV heads / head dim40 / 8 / 128
Context window131072
Modalitiesin: text; out: text
GGUFavailable
Commercial useyes

Minimum hardware (estimate)

At Q4_K_M and 8k context, total working set ≈ 9.3 GiB. On a machine with 64 GiB system RAM and no discrete GPU: CPU-only, slow. Limiting factor: No GPU VRAM; fits in system RAM but inference will be CPU-bound.

Open in fit advisor

Scores

BenchmarkValueUncertaintyRun bySource
GPQA Diamond29.9%±0.0210third-partyepoch-ai
MATH Level 510.8%±0.0061third-partyepoch-ai

Provenance

HF: mistralai/Mistral-Nemo-Instruct-2407 · snapshot 2026-08-10