Matching posts

Matching terms are highlighted

Building a Tesla Roadster 3D browser game with DeepSeek V4.1 Flash for ~$15 in API costs

A developer used DeepSeek V4.1 Flash via GMI Cloud to build a playable Tesla Roadster 3D browser game, spending about $15 in API costs.

deepseek-v4.1@lnkiai　Filed in No.020

1 day agoCopy original

POST

DeepSeek V4.1 Flash in Cursor: CursorBench 4.0 scores and per-task cost by thinking level

On CursorBench 4.0, DeepSeek V4.1 Flash scores 44.9% at $0.60 per task on Low up to 47.7% at $0.98 on Max, while Opus 5.5 Low gets 43.7% at $1.17.

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

11 hours agoCopy original

Using DeepSeek V4.1 Flash as an editor autocomplete model

The author wired DeepSeek V4.1 Flash into Wally as ghost-text autocomplete and says it is fast enough to write whole functions before the finger leaves Tab, with a demo video.

deepseek-v4.1@ShubhamMal72313　Filed in No.020

15 hours agoCopy original

DeepSeek V4.1 Flash in Claude Code: Rust coding hands-on

A user ran V4.1 Flash in Claude Code with rust-skills, reporting fast, cheap, clean Rust output and stable tool calling through long loops.

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

19 hours agoCopy original

DeepSeek V4.1 Flash 1017 on DeepSeek Harness: underrated but hallucinated a name

Testing DeepSeek V4.1 Flash 1017 in DeepSeek Harness, the author says the model is still underrated but was puzzled when it hallucinated a person's name that isn't theirs.

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

18 hours agoCopy original

DeepSeek V4.1 Flash Max vs GPT-6.1 Terra Ultra: Terra finishes in under 9 minutes, DeepSeek takes nearly an hour

A side-by-side test claims GPT-6.1 Terra Ultra finished in under 9 minutes with better quality, while DeepSeek V4.1 Flash Max took almost an hour.

deepseek-v4.1@SPAC89　Filed in No.020

17 hours agoCopy original

POST

DeepSeek V4.1 1017 wins on price for high-volume WhatsApp customer messages

The author compared eight models on small customer-message handling: GLM 5.3 Flash usable but slow, Haiku not smart enough, Sonnet/Opus too expensive, GPT 6.1 Terra weak, Grok 4.7 good but slow; the two Chinese models won on very low price and fast replies, while Opus 5.5 remains the pick for code.

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

15 hours agoCopy original

POST

Hands-on: DeepSeek V4.1 Flash 10-17 feels near Opus 4.6–4.8 but is very slow

A user says V4.1 Flash 10-17 performs around Opus 4.6–4.8 level but is very slow and makes worse guesses and theories than Claude.

deepseek-v4.1@mjkzys　Filed in No.020

14 hours agoCopy original

Hands-on: MiniMax M3.1 Flash codes an open-world FPS game over long sessions

The author ran M3.1 Flash via API in DeepSeek Harness and mcode over long sessions, adding features one by one to build a code-only open-world FPS with no external assets, calling it a big step up from M3 and planning to spend the rest of the month's quota improving it.

minimax-m3.1@ItsmeAjayKV　Filed in No.021

23 hours agoCopy original

POST

GTA one-shot test: V4.1 Flash 50 min for $0.36 vs Grok 4.7 $3.58

On the same one-shot GTA task, DeepSeek V4.1 Flash 1017 took 50 minutes for $0.36, Grok 4.7 took 41 minutes for $3.58, and Opus 5.5 was faster at 30 minutes; Flash is slow but very cheap.

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

8 hours 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.