Matching posts

Matching terms are highlighted

Fuse weekly update: model-agnostic agents and Fuse Compute

Fuse shipped a week of updates: agents switch between Claude and Codex with one config line, Fuse Compute offers dedicated NVIDIA GPUs in UK data centres, and its AI reviewer can approve select PRs without humans.

agent@UnsettledGreg　Filed in No.004

4 days agoCopy original

Solana weekly roundup: Altitude MCP v1 and more launches

A weekly Solana ecosystem roundup includes Altitude's MCP v1 connecting AI agents to transaction data, Flovia402's free AI API access, and brij's flight search APIs for agents.

agent@SolanaHub_　Filed in No.004

8 days agoCopy original

POST

Oct 20 AI brief: ChatGPT Slides collaboration, Runway MCP in Grok Bot

ChatGPT Slides adds real-time collaborative editing, Grok Bot gains Runway MCP, and new models including MiniMax M3.1 Flash and Mercury 3 launch the same day.

openai@testingcatalog　Filed in No.004

8 days agoCopy original

POST

Claude Duet should adopt a two-way product feedback channel

The author wants Anthropic's Claude Duet to offer product updates where users can ask questions and file complaints, contrasting it with OpenAI Dots.

personal-agent@vikktorrrre　Filed in No.012

20 hours agoCopy original

Grok Bot adds Pages

Grok Bot now has a Pages feature; the post shows only an image with no further detail on what it does.

grok@techdevnotes　Filed in No.012

5 days agoCopy original

Qoder extends free DeepSeek V4.1 Flash 1017 access to Nov 1

Qoder's international individual users can use DeepSeek V4.1 Flash 1017 free until Oct 31 midnight at a 0.0x billing rate, with a daily usage cap and no Teams or Enterprise access.

deepseek-v4.1@xunfenghellolo　Filed in No.020

21 hours agoCopy original

Strata 0.2.7: DeepSeek V4.1 Flash local prompt processing up to +65%

Strata 0.2.7 speeds up local DeepSeek-V4.1-Flash-1017 inference with up to 65% faster prompt processing, 47% on RTX 5070 at 4K, and native Strix Halo support reaching 300+ tok/s at 4K–32K.

deepseek-v4.1@NFT_Chen　Filed in No.020

19 hours agoCopy original

Strata 0.2.7: MoE disk-IO rework lets 12GB GPUs run DeepSeek V4.1 Flash locally

Strata 0.2.7 reworks expert-weight loading and adds adaptive attention buffering, claiming 47% faster 4K prompts and 33% faster 8K prompts on an RTX 5070 12G and 342 tok/s peak prompt reads on Strix Halo.

deepseek-v4.1@xiax0603　Filed in No.020

19 hours agoCopy original

POST

Strata v0.2.7: DeepSeek V4.1 Flash on Strix Halo, 512K/1M context

Strata 0.2.7 ships a dedicated gfx1151 engine package so DeepSeek V4.1 Flash runs on Ryzen AI Max Windows PCs, speeds up prompt reads (8K prompt 342 to 399 tok/s on four R9700 cards), and exposes 512K/1M context options.

deepseek-v4.1@uctive　Filed in No.020

21 hours agoCopy original

POST

DeepSeek site shows another deepseek-flash backend fingerprint change

The poster notes a new backend fingerprint change for deepseek-flash, recalling that an Oct 17 change preceded the V4.1 Flash 1017 release by hours, while cautioning it could just be a redeploy.

deepseek-v4.1@plotarmordev　Filed in No.020

1 day agoCopy original

Showing10of60posts

10 per batch

Keep scrolling to reveal posts 11–20

Showing the first 60 matches. Narrow the query with a model name or tag to find more specific results.