Qwen3.8-27B-MLX-6bit vs reka-edge-2603
Specs, VRAM requirements and download trends — updated 24 October 2026.
lmstudio-community · A · Open weights
Params
6.4 B
Context
—
30d
3.6 M
apache-2.0 · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | Qwen3.8-27B-MLX-6bit | reka-edge-2603 |
|---|---|---|
| Deployment | Open weights | Open weights · also via API |
| Parameters | 6.4 B | 7.1 B |
| Architecture | Qwen3_5ForConditionalGeneration | Yasa2ForConditionalGeneration |
| Context length | — | 16,384 |
| Licence | apache-2.0 | other |
| Commercial use | Allowed | Unknown |
| Languages | — | — |
| Downloads 30d | 3,566,589 | 845 |
| Downloads all time | 9.6 M | 18 K |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2026-08-14 | 2026-03-11 |
| API input price / 1M tokens | not on OpenRouter | $0.10 |
| API output price / 1M tokens | not on OpenRouter | $0.10 |
| Est. cost per 1M input + 1M output tokens | not on OpenRouter | $0.20 |
| 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 | 14,745 |
| API providers | not on OpenRouter | 1 |
| Tool calling | not on OpenRouter | Yes |
| Reasoning mode | not on OpenRouter | Yes |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
reka-edge-2603
$0.20
Download trend
Daily snapshots, last 63 days (23 Aug – 24 Oct)
4.7 M747
Qwen3.8-27B-MLX-6bit
reka-edge-2603
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
reka-edge-2603 · fp16
17.3 GB / 24 GB
Qwen3.8-27B-MLX-6bit · fp16
26.5 GB / 24 GB