DeepSeek V4 previews push frontier-level open weights at unusually low cost

// related

• #0001

  #0001Models & APIr/LocalLLaMA5 hours ago

  `Qwen3.7 Max` hits 5th on Artificial Analysis; 27B/35B still pending

  60radar

  Qwen3.7 MaxLarge language model — high-end Alibaba Qwen variant

  Artificial Analysis puts it near GPT 5.4 xhigh and above Gemini 3.5 Flash. Strong benchmark signal, but migration waits on API price and smaller-model results.

  • —Ranked 5th on Artificial Analysis, roughly tied with GPT 5.4 xhigh; credible enough to add to model eval lists.
  • —Gemini 3.5 Flash sits one step lower in the cited ranking, so latency/price will decide the practical winner.
  • —Qwen3.6 27B trails Max by 6 points; the 27B/35B Qwen3.7 results matter for local or cheaper deployment.

  Source: www.reddit.com/r/LocalLLaMA/comments/1tie6gy/qwen37_max_Read original →
• #0002

  #0002Models & APISimon Willison13 hours ago

  `llm-gemini` `0.32` adds `gemini-3.5-flash`

  50radar

  llm-geminiLLM CLI plugin — calls Gemini models from `llm`

  Simon Willison's llm CLI can now call Google's new Flash model through the Gemini plugin. Small update, but useful if your scripts already depend on llm.

  • —llm-gemini now exposes gemini-3.5-flash, so existing llm CLI workflows can test the model without custom API glue.
  • —Scope is one model alias in a plugin release. This is practical plumbing, not a new app-building capability by itself.
  • —Best fit is quick model comparison for summaries, extraction, and generation jobs already wired around Simon Willison's llm ecosystem.

  Source: simonwillison.net/2026/May/19/llm-gemini-2/#atom-everythRead original →
• #0003

  #0003Models & APISimon Willison14 hours ago

  `Gemini 3.5 Flash` ships broadly with a 3-6x price jump

  90radar

  Google is putting the new default-grade model into Search, Gemini, Android Studio, and the API. The API math changed: high-output features need a fresh margin check.

  • —Model ID is gemini-3.5-flash; it supports 1,048,576 input tokens and 65,536 output tokens. Strong fit for long-context document flows.
  • —Pricing is $1.50/M input and $9/M output, 3x 3 Flash Preview and 6x 3.1 Flash-Lite. Flash is no longer the obvious cheap default.
  • —Interactions API is in beta with server-side history management, echoing the Responses API pattern. Agent backends may get simpler state handling.
  • —No computer-use feature in this release. If your workflow depends on browser/desktop control, this is a model upgrade, not a full agent-runtime replacement.
  • —3.5 Pro is slated for next month, likely pricier. Build model routing now instead of hard-coding one Gemini tier.

  Source: simonwillison.net/2026/May/19/gemini-35-flash/#atom-everRead original →