Harness comparisonAI Coding
One reef prompt in three labs’ agents — DeepSeek Harness (DeepSeek V4.1 Flash 1017) vs OpenAI Codex (GPT-6 Astra) vs Claude Code (Claude Fable 5.1)
Scoreboard3
- DeepSeek V4.1 Flash 1017DeepSeekDeepSeek Harness
- Time
- 6.5 min
- Cost
- $0.09
- medium confidence
- Written “DeepSeek V4.1 Flash” after the 1017 update shipped; read as V4.1 Flash 1017.
- GPT-6 AstraOpenAIOpenAI Codex
- Time
- 2.1 min
- Cost
- $0.43
- Claude Fable 5.1AnthropicClaude Code
- Time
- 12 min
- Cost
- $6.00
Prompt
One reef prompt in three labs’ agentsThe post doesn’t quote the prompt wording.
DeepSeek V4.1 Flash 1017 in DeepSeek Harness, GPT-6 Astra in Codex and Claude Fable 5.1 in Claude Code get the same reef prompt, one run each.
Each side pairs a coding harness with its own model, so harness and model differences are mixed. More Claude Code vs OpenAI Codex. More OpenAI Codex vs DeepSeek Harness. More Claude Code vs DeepSeek Harness. All harness comparisons.
The post, itemised
- Task
- One reef prompt in three labs’ agents
- Category
- 3D & Shaders
- Mode
- Not stated
- Posted
- Video length
- 16 s
- Likes
- 0
- Views
- 70
- Bookmarks
- 0
- Sponsored
- No
NotesNo winner named; the author asks which reef you would ship. The video doesn’t label the panels.
Cost & time, as stated in the postAs stated: DeepSeek V4.1 Flash 1017 in DeepSeek Harness 6 min 32 s, $0.09; GPT-6 Astra in Codex 2 min 7 s, $0.43; Claude Fable 5.1 in Claude Code 12 min, $6.00.
Related
Posted by @agentsky_dev on X. aivs.tv claims no ownership: it hosts a copy for fast playback and links to the post on X. Request removal


