Qwen3 30B A3B
MoE
enthusiast
30.5B total params, 3B active per forward pass (MoE). Min memory based on full weight loading ~15GB.
AI-generated content marks
This model embeds text watermarks in generated text and adds C2PA provenance metadata to supported files such as .png, .jpg, and .svg.
Marks can be lost through editing, screenshots, or format conversion, so their absence does not prove a file is human-made.
Provider transparency docs →
coding
reasoning
- Parameters
-
30.5B
- Context
- 128k
- License
- apache 2.0
- Origin
-
🇨🇳
China
- Released
- Apr 2025
Run it locally
Per-quant memory needs and a static "can you run it?" reference - no rig entry required
Q4_K_M
15.5GB
15.5GB min
17.0GB rec
Balanced - the usual local sweet spot
Q8_0
31.0GB
31.0GB min
35.0GB rec
Near-lossless
The reference hardware
22 reference configs, drawn in-house. Scroll for more.
Can you run it? - reference rigs
| Rig |
Q4_K_M |
Q8_0 |
| NVIDIA Jetson Orin NX 16GB |
tight
|
no -> cloud
|
| Single GTX 1080 Ti (11GB) |
offload
|
no -> cloud
|
| 4x H100 80GB (320GB) |
fast 4827.1t/s
|
fast 2415.3t/s
|
| NVIDIA DGX Station 748GB |
fast 2881.9t/s
|
fast 1442.0t/s
|
| 8x RTX 3090 rack (192GB) |
fast 2698.0t/s
|
fast 1350.0t/s
|
| 4x RTX 5090 (128GB) |
fast 2582.2t/s
|
fast 1292.0t/s
|
| AMD Instinct MI300X (192GB) |
fast 1918.2t/s
|
fast 959.8t/s
|
| 4x RTX 4090 (96GB) |
fast 1452.5t/s
|
fast 726.8t/s
|
| 2x RTX 5090 (64GB) |
fast 1291.1t/s
|
fast 646.0t/s
|
| 2x RTX 3090 (48GB) |
fast 674.5t/s
|
fast 337.5t/s
|
| Single RTX 5090 (32GB) |
fast 645.5t/s
|
tight
|
| RTX PRO 6000 Blackwell (96GB) |
fast 645.5t/s
|
fast 323.0t/s
|
| Mac Studio M4 Ultra 192GB |
fast 429.1t/s
|
fast 214.7t/s
|
| Mac Studio M4 Ultra 512GB |
fast 429.1t/s
|
fast 214.7t/s
|
| Single RTX 4090 (24GB) |
fast 363.1t/s
|
offload
|
| MacBook Pro M5 Max 128GB |
fast 241.3t/s
|
fast 120.7t/s
|
| Dual EPYC 9004 + 768GB DDR5-4800 |
fast 166.0t/s
|
fast 83.1t/s
|
| DGX Spark 128GB unified |
fast 98.3t/s
|
fast 49.2t/s
|
| Ryzen AI Max+ 395 128GB |
fast 92.2t/s
|
fast 46.1t/s
|
| Jetson AGX Orin 64GB |
fast 73.8t/s
|
fast 36.9t/s
|
| Epyc + 512GB DDR4-3200 + 2x RTX 3090 |
fast 73.8t/s
|
fast 36.9t/s
|
| Epyc + 512GB DDR4-2400 + 2x RTX 3090 |
fast 55.3t/s
|
fast 27.7t/s
|
Fit tiers use the same will-it-run logic as the rig finder. For comfortable fits, the badge reflects decode speed: fast >=20 t/s, ok 8-20 t/s, slow <8 t/s. t/s is a bandwidth estimate, not a measured benchmark.
Or run it in the cloud
Live per-provider pricing, throughput and uptime - refreshed about 19 hours ago via OpenRouter. Click a column to sort.
|
Provider
|
Type
|
Input $/M
|
Output $/M
|
Cache $/M
|
Tok/s
|
Latency
|
Uptime
|
Value
|
|
DeepInfra
|
API
|
0.12 |
0.50 |
- |
- |
- |
100.00% |
best uptime
|
Default order: throughput among 95%+ uptime providers, then latency; subscriptions last. Sort by any column. Subscription rows show $/mo in the Value column - per-token columns are "-". Affiliate links are marked sponsored / nofollow. Confirm current pricing on the provider's site before committing.
PRICE HISTORY
Inference cost over time
Data accumulates from the first daily sync - longer ranges populate over time. Prices come from OpenRouter snapshots, not a historical API.
Loading price history...