The Open-Source AI Stack
RSS
All models

Models · mistral

Mistral 7B v0.1

Open Mistral AI · 2023-09-27 · Apache-2.0

The first open-weights model to beat Llama 2 13B at 7B scale using sliding-window attention. Mistral's Apache-2.0 release reset what "open" meant after Llama 2's community license.

Cost

— / Mtok input
— / Mtok output

Together AI · as of 2026-05-19

via Artificial Analysis ↗

Speed

— tok/sec output

Together AI · as of 2026-05-19

via Artificial Analysis ↗

Architecture

tokens in Embedding vocab 32,000 · mistral tokenizer × 32 layers GQA + Sliding-window (interleaved) RoPE context 8,192 tokens Dense MLP SwiGLU activation (standard) 7.2B active params Output projection tokens out
Schema-generated from data/models.yaml. Every label is auditable against the model's sources.

Specs

Architecture
dense
Total params
7B
Active params
7.2B
Context window
not verified
Attention
hybrid-gqa-sliding
Position encoding
rope
OSI-approved
yes
Data released
no
Training code
not released

Benchmarks

Each score carries the date it was published; we never infer or interpolate missing scores.

Recommended use cases

  • local deployment
  • Apache-2.0 reference
  • fine-tuning base

Available quantizations

GGUF llama.cpp's container; the common local format, k-quants from Q2 to Q8. runs on llama.cpp, Ollama
AWQ Activation-aware 4-bit weight quantization for GPU serving. runs on vLLM, SGLang
GPTQ Post-training 4-bit weight quantization for GPU serving. runs on vLLM, SGLang, Transformers
EXL2 ExLlamaV2's variable-bitrate format for consumer GPUs. runs on ExLlamaV2
MLX Apple MLX 4/8-bit layout for Apple silicon. runs on Apple MLX
FP8 8-bit float, frequently a native release on Hopper / Blackwell GPUs. runs on vLLM, SGLang, TensorRT-LLM
bitsandbytes On-the-fly NF4 / INT8 weight quantization inside Transformers. runs on Transformers

Verified via the Hugging Face model tree ↗. Community quantizations change over time; the families shown are those with published weights at audit time.

Notable innovations

  • · Sliding-window + GQA hybrid attention
  • · Apache-2.0 release

Lineage

Architecture base for the Mixtral MoE.

Derivatives

Sources