Llama-3.1-8B-Instruct vs Ministral-3-8B-Instruct-2512
Specs, VRAM requirements and download trends — updated 24 October 2026.
meta-llama · A · Open weights
Params
8.0 B
Context
—
30d
4.9 M
llama3.1 licence · conditional
mistralai · B · Open weights
Params
8.9 B
Context
—
30d
94 K
apache-2.0 · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | Llama-3.1-8B-Instruct | Ministral-3-8B-Instruct-2512 |
|---|---|---|
| Deployment | Open weights · also via API | Open weights · also via API |
| Parameters | 8.0 B | 8.9 B |
| Architecture | LlamaForCausalLM | Mistral3ForConditionalGeneration |
| Context length | 131,072 | 262,144 |
| Licence | llama3.1 | apache-2.0 |
| Commercial use | Conditional | Allowed |
| Languages | 8 | 11 |
| Downloads 30d | 4,864,172 | 93,976 |
| Downloads all time | 191.0 M | 1.2 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | Yes — accept licence | No |
| First seen on the Hub | 2024-07-18 | 2025-10-31 |
| API input price / 1M tokens | $0.05 | $0.15 |
| API output price / 1M tokens | $0.08 | $0.15 |
| Est. cost per 1M input + 1M output tokens | $0.13 | $0.30 |
| Cached input / 1M tokens | $0.025 | $0.015 |
| AA Intelligence Index | 6.9 | 5.5 |
| AA Coding Index | 5.4 | 9.7 |
| AA Agentic Index | — | 0.6 |
| Max output tokens | 117,964 | 209,715 |
| API providers | 5 | 1 |
| Tool calling | Yes | Yes |
| Reasoning mode | No | No |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
Llama-3.1-8B-Instruct
$0.13
Ministral-3-8B-Instruct-2512
$0.30
Llama-3.1-8B-Instruct is 2.3× cheaper at list price.
Download trend
Daily snapshots, last 89 days (28 Jul – 24 Oct)
8.0 M105 K
Llama-3.1-8B-Instruct
Ministral-3-8B-Instruct-2512
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
Llama-3.1-8B-Instruct · fp16
19.4 GB / 24 GB
Ministral-3-8B-Instruct-2512 · fp16
24.8 GB / 24 GB