v2.2 · 2026-06-12
Mistral Nemo 12B Instruct 2407
mistral-nemo-12b-instruct-2407
Mistral Nemo 12B Instruct 2407 is a 12B text chat model profile for local AI planning. This page records provider, license, context, quantization, hardware-fit estimates, setup hints, source links, and caveats so readers do not choose a model by name alone.
PROVIDER
Mistral AI / NVIDIA
FAMILY
mistral
MODEL TYPE
text-chat
PARAMETERS
12B
MODALITIES
text
ARCHITECTURE
decoder-only Transformer using the Mistral architecture
CONTEXT WINDOW
128,000 tokens
TRAINING TOKENS
—
RELEASE DATE
2024-07-18
GGUF REPOSITORIES
bartowski/Mistral-Nemo-Instruct-2407-GGUF community conversion
Community GGUF conversion with Q4_K_M, Q5_K_M and Q8_0 file-size table.
SETUP HINTS
OLLAMA
ollama run mistral-nemo:12bLM STUDIO
Use bartowski/Mistral-Nemo-Instruct-2407-GGUF and select Q4_K_M or Q5_K_M.
LLAMA.CPP
llama-server -hf bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_MQUANTIZATION FILE SIZES (GGUF)
Q4_K_M
7.48 GB
Q5_K_M
8.73 GB
Q8_0
13.02 GB
Quantization note: Q4_K_M
RAM / VRAM ESTIMATES
MIN RAM
12 GB
COMFORTABLE RAM
16 GB
MIN VRAM
10 GB
COMFORTABLE VRAM
12 GB
These are conservative local-inference estimates, not official hardware requirements.
HARDWARE FIT
CPU only (no GPU)
yes
8 GB RAM
no
16 GB RAM
comfortable
32 GB RAM
comfortable
8 GB VRAM
limited
12 GB VRAM
good
24 GB VRAM
comfortable
Apple Silicon (unified memory)
comfortable_on_16gb_or_more_unified_memory
Hardware fit values are conservative local-inference estimates based on GGUF size plus runtime and KV-cache overhead. Actual requirements depend on context length, quantization, runtime, GPU offload, and other running apps.
BEST FOR
·Apache-2.0 12B local assistant
·long-context local use up to 128k when memory permits
·coding and world-knowledge tasks in the 12B class
·drop-in upgrade path from Mistral 7B-class systems
AVOID IF
·you only have 8 GB RAM
·you need a tiny edge model
·you need verified explicit tool-call support from the fetched model card
CAVEATS
·RAM and VRAM estimates are conservative local-inference estimates based on GGUF file size plus runtime and KV-cache overhead, not official Mistral requirements.
·The fetched sources describe multilingual and code data but did not provide a verified exhaustive language list, so languages is empty.
·Explicit tool-call support was not independently verified in the fetched reliable sources; supportsTools is null.
SOURCE URLS
FIELD EVIDENCE
releaseDatemistral.ai/news/mistral-nemo
architecturemistral.ai/news/mistral-nemo
contextWindowTokenshuggingface.co/mistralai/Mistral-Nemo-Instruct-2407
CAPABILITIES
No special capabilities flagged.