Inkling 2 Mini on PC: edges out Qwen 3.8 27B on coding tests but painful to run
Ubunta · x · 2026-11-21
@Ubunta benchmarked Inkling 2 Mini locally on a Ryzen AI 9 HX 370 (Radeon 890M iGPU) and concludes the hype isn't justified on PC yet — maybe on Mac.
Code quality is real: it edged out Qwen 3.8 27B on his hard coding tests (277 vs 272 asserts) with cleaner output. But it wasn't faster (9.2–10.5 tok/s vs Qwen's 8–14 with Ollama MTP), used no fewer tokens, and required a forked llama.cpp plus unmerged PRs to get running. Qwen stays his daily driver; Inkling stays installed.
More from Infra
- Inkling 2 Mini edges Qwen 3.8 27B on coding (277 vs 272) but struggles on PC hardware — Ubunta · 2026-11-21
- Inkling-2-Mini runs on laptops with as little as 9GB RAM, per Unsloth local guide — NielsRogge · 2026-11-21
- Only when infra resources burn out will the application layer see its spring — yangyi · 2026-11-21
- 26 LLM routers caught injecting malicious tool calls, draining a client's $500k wallet — 0xkarasy · 2026-11-21
- OpenRouter accused of sending users' BYOK private keys to China despite earlier denials — 0xkarasy · 2026-11-21
- OVH's Olala Mini 7A1B: 360 tok/s decode on one RTX 5090, 96K context, v2.0 to exceed 100B — JiliJeanlouis · 2026-11-21