live · updates every second
Mid
$0.450
0% under hosted
Spread
$0.0000
0 bps
Last
—
no fills yet
Capacity
0 tok/s
0 nodes
▾
27B · Q4_K_M · 32.8K token context · per Mtok output
no fills in the last 1m window
Watchlist
29/29
- Mistral Medium 3.5 128B128B——
- GPT-OSS 120B120B——
- Llama 3.3 70B Instruct70B——
- Laguna XS 2.1 33B A3B33B-A3B——
- Qwen2.5 Coder 32B32B——
- Qwen3 32B32B——
- Qwen3 30B A3B30B-A3B——
- Qwen3 Coder 30B30B-A3B——
- Muse Glimmer 30B30B——
- Nemotron 3.5 Lightning 30B A3B30B-A3B——
- Gemma 2 27B IT27B——
- Qwen3.5 27B27B——
- Gemma 4 26B26B——
- Mistral Small 24B24B——
- GPT-OSS 20B20B——
- DeepSeek R1 Distill 14B14B$0.3750.0%
- Qwen3 14B14B$0.3750.0%
- Gemma 4 12B12B$0.350▼-30.0%
- Llama 3.1 8B Instruct8B$0.160▲+900.0%
- Qwen3 8B8B$0.010▼-96.7%
- LFM2.5 8B A1B8B-A1B——
- Granite 4.1 8B8B$0.300▼-40.0%
- Qwen3.5 4B4B$0.1700.0%
- Phi-3.5 Mini Instruct3.8B$0.1250.0%
- Llama 3.2 3B Instruct3B$0.090▼-20.0%
- Qwen2.5 3B Instruct3B$0.1130.0%
- Gemma 2 2B IT2B$0.0750.0%
- Llama 3.2 1B Instruct1B$0.0500.0%
- Qwen2.5 0.5B Instruct0.5B$0.0300.0%
Order book
No resting orders on this book yet.
Depth
Cumulative resting size by price
No resting depth on this side yet.
Recent fills
Coloured by aggressor side
PriceSizeTime (UTC)
Nodes quoting this model
0 online
| Node | GPU | Ask | Tok/s | Ceiling |
|---|
How this book matches
Rules applied in priority order
- 1Physical eligibility
A node only sees the queue for models its VRAM and memory bandwidth can actually serve at its quoted speed. Everything below runs on that eligible set.
- 2Price priority
The cheapest eligible ask fills first. Nothing outranks price — a better quote always jumps the line.
- 3Time breaks ties
Equal asks fill oldest-first, so posting early is rewarded but never beats a lower price.
- 4Uptime is the last tie-break only
Ranking by session length alone would hand every job to the single longest-lived node and starve new supply, so it settles ties and nothing more.
Book snapshot 2026-09-13 04:32:13 UTC.