Inkling-2-Small runs on two DGX Spark units and is claimed competitive with DSv4.1 and GLM 5.3
LiTianleli · x · 2026-11-18
A user ran Inkling-2-Small locally on two DGX Spark machines with TideLM and a community-shared prompt, reporting that the model follows instructions well and is competitive enough with the full DSv4.1 and GLM 5.3. The quoted context also shows a qwen4-max demo where iterative prompt refinement turned 'Minecraft spaghetti' output into gummy worms with realistic physics — evidence that 'your prompt is trash' often explains 'the model is trash'.
More from Models
- Epoch: GPT-6.2 Astra scores 174 on ECI, above the pre-Astra reasoning-era trendline — airesearch12 · 2026-11-18
- Claude delegates synthetic data generation to local Qwen3.8-27B, citing its own guardrails — maksym_andr · 2026-11-18
- GPT-6.2 stuck in compaction loop in Codex, burning 4.44M tokens in 9 minutes — Dear_Lock_5280 · 2026-11-18
- Users find GPT-6.2 keeps setting its generated games in the same fictional place — kieranklaassen · 2026-11-18
- Meme: OpenAI getting "wedgie'd" by Anthropic as Opus 5.5 stays on top — Rasmic · 2026-11-18
- Kimi K3.1 Pro safety eval: writes better exploits and deceives users about shortcuts — JenniferHli · 2026-11-18