Public preview · figures are illustrative fixture data, attributed to their source · data to May 2025 · read-only
Model · Mixtral 8x7B · December 2023 Live · Hugging Face

Mixtral 8x7B

A 47 billion-parameterHow many numbers the model learned during training — the usual rough measure of its size. mixture-of-expertsA model split into parts where only a few run for each word, so it answers faster than its size suggests. model using ~13 billion parameters per token from Mistral AI. Sparse MoE with 8 experts, 2 active per token.

Size
46.7B12.9B active
Memory to run
~30 GBsmallest, 8K ctx
Context
32Ktokens
License
PermissiveApache License 2.0
Updated
13 Sept 2026

Which version to use

Where it runs

Best-fitting version on each reference system, at an 8K contextHow much text the model can consider at once, counted in tokens — roughly ¾ of a word each..

Runs well 9

Slowly (CPU or offload) 4

Too large 0

None of the reference systems.

Benchmarks

Each benchmarkA fixed set of questions every model is given, so their scores can be compared on the same task. is developer-reported; prompts and settings differ between labs. Bars are relative to the best open result.

No benchmark results recorded for this model.

Variants & downloads

Each variant is a separate set of weights. Expand one for its downloads: quantizationStoring each of the model’s numbers with fewer bits, so the file is smaller and needs less memory. trades a little quality for a much smaller file, and the memory column adds the working memoryExtra space the model needs while it answers. It grows with the length of the conversation, on top of the file itself. a conversation needs on top.

Variant Mixtral 8x7B Instruct v0.1instruct · Mistral AI · 2 downloads · Apache License 2.0Permissive
What it is
Tuned to follow instructions and hold a conversation. The usual choice.
Publisher
Mistral AI
License
Apache License 2.0 · commercial use allowed
Released
11 Dec 2023

Contributions are not open yet, so there is nothing here from members.

QuantizationFormatBitsDownloadMemory @ 8KPublisher
Download BF16nativesafetensors1687.0 GiB~92.0 GiBMistral AImistralai/Mixtral-8x7B-Instruct-v0.1
Download Q4_K_Mk quantgguf4.8926.6 GiB~29.2 GiBLM Studio Communitylmstudio-community/Mixtral-8x7B-Instruct-v0.1-GGUF
Variant Mixtral 8x7B Instruct v0.1 Parasitefine tune · ApolloRaines (third-party) · 0 downloads · Apache License 2.0Permissive
What it is
A community or third-party fine-tune of another variant.
Publisher
ApolloRaines
License
Apache License 2.0 · commercial use allowed
Released
21 Jul 2026

Contributions are not open yet, so there is nothing here from members.

No downloadable artifacts recorded.

Architecture detailsLayers, attention and KV cache geometry
Parameters
46.70B
Active / token
12.9B
Architecture
Mixture of experts
Layers
32
Attention heads
32
KV heads
8
Head dim
128
KV cache @ 8K (fp16)
1.00 GiB
Max context
32,768 tokens

Measured performance

Throughput on specific systems and runtimes, with the source of each measurement.

No performance measurements recorded yet.

Community results

Contributions are not open yet, so there is nothing here from members.

Reviews

Contributions are not open yet, so there is nothing here from members.

Lineage

How this model’s variants relate to each other and to other models.

VariantMixtral 8x7B Instruct v0.1 Parasite fine-tuned from Mixtral 8x7B Instruct v0.1

Sources & history

Sources

  • Hugging Face Hub (live source, 1 record, 13 Sept 2026)

External identifiers

Field history

  • base = "mistralai/Mixtral-8x7B-Instruct-v0.1" Hugging Face Hub (current)
  • derivation = "fine_tune" Hugging Face Hub (current)
  • license = "apache-2.0" Hugging Face Hub (current)
  • name = "Mixtral 8x7B Instruct v0.1 Parasite" Hugging Face Hub (current)