opus-mt-en-fr vs Hy-MT2-30B-A3B
Specs, VRAM requirements and download trends — updated 10 November 2026.
Helsinki-NLP · A · Open weights
Params
—
Context
512
30d
192 K
apache-2.0 · commercial OK
tencent · B · Open weights
Params
30.1 B
Context
262,144
30d
95 K
apache-2.0 · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | opus-mt-en-fr | Hy-MT2-30B-A3B |
|---|---|---|
| Deployment | Open weights | Open weights · also via API |
| Parameters | — | 30.1 B |
| Architecture | MarianMTModel | HYV3ForCausalLM |
| Context length | 512 | 262,144 |
| Licence | apache-2.0 | apache-2.0 |
| Commercial use | Allowed | Allowed |
| Languages | — | 36 |
| Downloads 30d | 192,334 | 94,727 |
| Downloads all time | 18.1 M | 201 K |
| Quantizations on the Hub | — | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2022-03-02 | 2026-05-11 |
| API input price / 1M tokens | not on OpenRouter | $0.074 |
| API output price / 1M tokens | not on OpenRouter | $0.30 |
| Est. cost per 1M input + 1M output tokens | not on OpenRouter | $0.37 |
| Cached input / 1M tokens | not on OpenRouter | — |
| AA Intelligence Index | not on OpenRouter | — |
| AA Coding Index | not on OpenRouter | — |
| AA Agentic Index | not on OpenRouter | — |
| Max output tokens | not on OpenRouter | 4,096 |
| API providers | not on OpenRouter | 1 |
| Tool calling | not on OpenRouter | No |
| Reasoning mode | not on OpenRouter | No |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
Hy-MT2-30B-A3B
$0.37
Download trend
Daily snapshots, last 90 days (13 Aug – 10 Nov)
229 K95 K
opus-mt-en-fr
Hy-MT2-30B-A3B
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
Hy-MT2-30B-A3B · fp16 @ 256K ctx
210.9 GB / 24 GB