Creeta

Creeta

Jun 18

Claude Code v2.1.156: Opus 4.8 Thinking Block Hotfix Explained

#claudecode #opus48 #hotfix #extendedthinking

8 min read

Creeta

Jun 18

langchain-fireworks 1.4.2: Annotated + ChatFireworks Quickstart

#langchain #fireworksai #chatfireworks #llmframework

7 min read

Creeta

Jun 18

openai-codex v0.1.0b1: First Beta Install and Thread Walkthrough

#openaicodex #python #sdk #beta

7 min read

Creeta

Jun 18

'Gemini Omni 3.5' doesn't exist. Here's the real split.

#geminiomni #videogeneration #googleio2026 #tutorial

7 min read

Creeta

Jun 18

You don't pick the RL algorithm — SIA's Feedback loop does

#sia #selfimprovingai #hexolabs #lora

8 min read

Creeta

Jun 18

Qwen3.6-35B NVFP4 runs on one H100 — A100 owners are out

#qwen3 #nvfp4 #vllm #nvidia

8 min read

Creeta

Jun 18

Step 3.7 Flash is a drop-in — except for one endpoint detail

#stepfun #step37flash #llm #moe

1

9 min read

Creeta

Jun 18

llama-bench skipped FA on capable GPUs — b9437 corrects it

#llamacpp #llm #gguf #flashattention

7 min read

Creeta

Jun 18

Opus 4.8 kills budget_tokens — here's what else moved

#claude #opus48 #anthropic #llm

7 min read

Creeta

Jun 18

Composer 2.5 hits near-frontier at 60 lower spend

#cursor #composer25 #codingagent #kimik2

7 min read

Creeta

Jun 18

Nemotron 3 Ultra went live June 4. Here's the call that works.

#nvidia #nemotron #openweights #llm

7 min read

Creeta

Jun 18

Windsurf is Devin Desktop now. Cascade has 27 days left.

#devin #cognition #windsurf #devindesktop

7 min read

Creeta

Jun 18

Is Omni's conversational video editor as good as the demos?

#geminiomni #googleflow #videoai #multimodal

1

7 min read

Creeta

Jun 18

NeMo out, GGUF in: how parakeet.cpp ports NVIDIA ASR to C++

#parakeet #ggml #gguf #asr

6 min read

Creeta

Jun 18

Qwen3 in the browser, zero keys — WebLLM 0.2.83 hands-on

#webllm #webgpu #browserai #ondeviceinference

7 min read

Creeta

Jun 18

NVIDIA's 550B finally lands: free to use, expensive to host

#nvidia #nemotron #llm #mixtureofexperts

6 min read

Creeta

Jun 18

MiniMax M3 benchmarks at $0.30/M: verified vs. vendor-only

#minimax #minimaxm3 #opensource #sparseattention

7 min read

Creeta

Jun 18

Meta Business AI went global — gated rollout, paid plans TBD

#meta #whatsapp #businessai #messenger

6 min read

Creeta

Jun 17

Gemma 4 12B skips the audio encoder. Is 16 GB enough?

#gemma #gemma4 #google #openweight

7 min read

Creeta

Jun 17

K2.7 Code is 30% lighter — but chain-of-thought is locked on

#kimi #moonshotai #k27code #cli

7 min read

Creeta

Jun 17

A 10-second Grok Imagine 1.5 clip at 720p runs $1.41

#grok #xai #imagetovideo #videogeneration

5

7 min read

Creeta

Jun 17

Ideogram 4 was trained on JSON — plain prompts are second-class

#ideogram #ideogram4 #texttoimage #jsonprompting

7 min read

Creeta

Jun 17

macOS 27 skips Intel — Siri AI is queued, Liquid Glass is not

#ios27 #macos27 #wwdc2026 #goldengate

6 min read

Creeta

Jun 17

Hand off any session to the GUI — Codex CLI 0.138.0

#openai #codex #codexcli #01380

5 min read

Creeta

Jun 17

Kog hits 3K t/s on MI300X, no kernel switches — test it now

#amd #mi300x #kogai #monokernel

1

8 min read

Creeta

Jun 17

Fable 5 refusals are 200 OK — your error handler misses them

#claudefable5 #anthropic #llm #apiintegration

1

7 min read

DEV Community

Badges

Writing Debut

Claude Code v2.1.156: Opus 4.8 Thinking Block Hotfix Explained

langchain-fireworks 1.4.2: Annotated + ChatFireworks Quickstart

openai-codex v0.1.0b1: First Beta Install and Thread Walkthrough

'Gemini Omni 3.5' doesn't exist. Here's the real split.

You don't pick the RL algorithm — SIA's Feedback loop does

Qwen3.6-35B NVFP4 runs on one H100 — A100 owners are out

Step 3.7 Flash is a drop-in — except for one endpoint detail

llama-bench skipped FA on capable GPUs — b9437 corrects it

Opus 4.8 kills budget_tokens — here's what else moved

Composer 2.5 hits near-frontier at 60 lower spend

Nemotron 3 Ultra went live June 4. Here's the call that works.

Windsurf is Devin Desktop now. Cascade has 27 days left.

Is Omni's conversational video editor as good as the demos?

NeMo out, GGUF in: how parakeet.cpp ports NVIDIA ASR to C++

Qwen3 in the browser, zero keys — WebLLM 0.2.83 hands-on

NVIDIA's 550B finally lands: free to use, expensive to host

MiniMax M3 benchmarks at $0.30/M: verified vs. vendor-only

Meta Business AI went global — gated rollout, paid plans TBD

Gemma 4 12B skips the audio encoder. Is 16 GB enough?

K2.7 Code is 30% lighter — but chain-of-thought is locked on

A 10-second Grok Imagine 1.5 clip at 720p runs $1.41

Ideogram 4 was trained on JSON — plain prompts are second-class

macOS 27 skips Intel — Siri AI is queued, Liquid Glass is not

Hand off any session to the GUI — Codex CLI 0.138.0

Kog hits 3K t/s on MI300X, no kernel switches — test it now

Fable 5 refusals are 200 OK — your error handler misses them