Mistral-Large-4-1T-A52B
Mistral Large 4 is our largest and most capable model to date: a 1 trillion-parameter natively multimodal Mixture-of-Experts model with 49 billion active parameters per token (52 billion including embeddings and output layers), trained from scratch on 3,800 NVIDIA Grace Blackwell GPUs in Mistral's own datacenters in Europe.
Params
52.0 B
Context
—
Downloads 30d
12 K
Likes
1,644
Download history
daily snapshots · 11 days12 K278
Oct 31Nov 3Nov 7Nov 10
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| model.safetensors | fp16 | 1,060.6 GB | 1,174.9 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Use via API
via OpenRouter · USD per 1M tokens · synced 2026-11-10$0.70
input / 1M
$2.10
output / 1M
$0.07
cached input / 1M
$2.80
1M in + 1M out
- Context window
- 1,048,576 tokens
- Max output
- 262,144 tokens
- Providers
- 1
- Cheapest provider
- Mistral · $0.70 / $2.10
- Best throughput
- —
- AA Intelligence / Coding / Agentic
- 26.1 / — / —
✓ Tool calling
✓ Reasoning
✓ Structured outputs
✓ Image input
✗ File input
✗ Audio input
✓ Prompt caching
Specifications
- Architecture
- MistralLarge4ForConditionalGeneration
- Parameters
- 52.0 B
- Licence
- other
- First seen on the Hub
- 2026-09-30
- Training datasets
- undisclosed
- Added to our catalog
- 2026-10-31
Compare with any image-text-to-text model
Popular comparisons
Appears in