Matching posts

Matching terms are highlighted

POST

User reports DeepSeek V4.1 Flash 1017 takes ~65 minutes per run on vLLM

A user says a DeepSeek V4.1 Flash 1017 run on their work machine takes about 65 minutes, and notes vLLM is fast because it tunes for the GPU and precompiles CUDA graphs before serving.

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

1 day agoCopy original

POST

User calls DeepSeek V4.1 a steal for running a Hermes agent

A user says DeepSeek V4.1 is a steal for running a Hermes agent, with an Opencode Go $10 subscription giving near-unlimited capacity and direct access saving more.

deepseek-v4.1@brodock　Filed in No.020

22 hours agoCopy original

POST

User: DeepSeek V4.1 Flash still best open model at its size for overnight agent loops

A brief usage report says deepseek-v4.1-flash-1017 remains the strongest open model at its size for agent loops running overnight.

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

8 hours agoCopy original

POST

Agent user: DeepSeek's stronger, shorter, cheaper output makes the bill matter more than benchmarks

An all-day agent operator says DeepSeek's newer model is stronger, answers more concisely and costs less, and that for heavy agent users the actual bill matters more than benchmark scores.

deepseek-v4.1@pranavbhatia123　Filed in No.020

1 day agoCopy original

Claude Duet update ships user-requested feedback

A user says the latest Claude Duet app update notes include feedback they submitted the same day, pointing to fast iteration.

personal-agent@mrfundman　Filed in No.012

2 days agoCopy original

Google Flow feedback: too few free credits to test prompts

A user says Flow's free credits ran out after the first trial and asks for more so creators can test prompts and refine results before upgrading.

google-flow@ellabarbie_　Filed in No.017

3 days agoCopy original

POST

T3 Code misses Gemini 4 Argon until ACP runtime reinstall

A user trying T3 Code found it only listed Flash and stale cached models instead of Gemini 4 Argon; reinstalling the ACP runtime fixed detection, and they may switch to it full-time.

gemini-4@thtbee_　Filed in No.008

2 days agoCopy original

POST

User take: DeepSeek V4.1 Flash is the best model to hold a conversation with

From limited input the author finds V4.1 Flash the best conversational model, agreeing that GPT-6.1 Sol often sounds forced and trained to be relatable.

deepseek-v4.1@st3llarorb　Filed in No.020

3 days agoCopy original

POST

DeepSeek V4.1 Flash slowdown reports traced to harness differences

The author argues the API did not get slower: the same model finishes tasks at different speeds across harnesses, with Pi Agent faster than DSH, creating a false impression.

deepseek-v4.1@Suercha007　Filed in No.020

3 days agoCopy original

Jio Pro users report Gemini 4 Argon still unavailable in Antigravity

A user says Jio Pro is supposed to unlock Gemini 4 Argon and Opus 5.5 in Antigravity, but it still doesn't work for them and asks if anyone has access.

gemini-4@xKaka1456　Filed in No.008

4 days 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.