Mixtral-8x22B-Instruct-v0.1
py from mistralcommon.tokens.tokenizers.mistral import MistralTokenizer from mistralcommon.protocol.instruct.messages import UserMessage from mistralcommon.protocol.instruct.request import ChatCompletionRequest
Params
140.6 B
Context
65,536
Downloads 30d
46 K
Likes
777
Download history
tracking started — chart appears after 7 days of snapshots (5 recorded)46 K downloads in the last 30 days
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 | bf16 | 281.3 GB | 331.0 GB | ❌ Won’t fit |
| model.safetensors (bf16, full) | bf16 + 64K ctx | 281.3 GB | 478.6 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ curl -s https://aimodelscomparison.com/api/v1/models/mixtral-8x22b-instruct-v0-1
{
"hf_id": "mistralai/Mixtral-8x22B-Instruct-v0.1",
"params_b": 140.63,
"context_length": 65536,
"license": { "id": "apache-2.0", "commercial": "yes" },
"downloads_30d": 46472,
"vram_estimates": [
{ "quant": "bf16", "gb": 331.0 }
],
"updated_at": "2026-10-19T06:01:47Z"
}
Use via API
via OpenRouter · USD per 1M tokens · synced 2026-10-24$2.00
input / 1M
$6.00
output / 1M
$0.20
cached input / 1M
$8.00
1M in + 1M out
- Context window
- 65,536 tokens
- Max output
- 52,428 tokens
- Providers
- 1
- Cheapest provider
- Mistral · $2.00 / $6.00
- Best throughput
- —
- AA Intelligence / Coding / Agentic
- 5.7 / — / —
✓ Tool calling
✗ Reasoning
✓ Structured outputs
✗ Image input
✓ File input
✗ Audio input
✓ Prompt caching
Specifications
- Architecture
- MixtralForCausalLM
- Parameters
- 140.6 B
- Tensor type
- BF16
- Context length
- 65,536
- Vocabulary
- 32,768
- Layers / heads
- 56 / 48
- Licence
- apache-2.0
- First seen on the Hub
- 2024-04-16
- Training datasets
- undisclosed
- Added to our catalog
- 2026-10-20
Compare with any text-generation model