@LeonardMainnet

data pipelines @datacurve

Joined March 2014
  • Tweets2.4K
  • Following1.5K
  • Followers4.9K
272Media
Leonard retweeted
Wenqi/Kevin
@winkey_hAug 21
Ok ox-alpha has that big model smell I'm getting around 63% at 47K avg output tokens on a DeepSWE subset This is pareto optimal amongst open models and just shy of Grok 4.6 compared to closed models!
OpenCode
@opencodeAug 20
Ox Alpha (stealth model) is free for the next week - 1M Context - Multi-modal - Zero Data Retention Generous rate limits, near unlimited usage We have capacity for 100T tokens per day, lets see what you can do
32
26
17
559
196,644
Leonard retweeted
Datacurve
@datacurveAug 13
Gemini 3.7 Flash debuts at 65.5% on DeepSWE. It delivers substantial improvements over 3.6 Flash, scoring +18.8% higher while costing less than half as much per task.
47
53
30
799
225,472
Leonard retweeted
Datacurve
@datacurveJul 22
Gemini 3.6 Flash is now on DeepSWE at 49%, roughly on par with Opus 4.8 at medium reasoning.
51
51
15
1,314
249,683
Leonard retweeted
Datacurve
@datacurveJul 17
Kimi K3 debuts at #3 on DeepSWE. It's the first open-weights model that delivers frontier-level performance, achieving results similar to Claude Fable and GPT-5.6 Sol.
106
339
127
3,652
670,342
Leonard retweeted
Datacurve
@datacurveJul 9
GPT-5.6 tops the DeepSWE leaderboard at 73%. Sol, Terra, and Luna results are now available.
71
168
88
2,031
347,934
Leonard retweeted
Datacurve
@datacurveJun 19
Claude Fable 5 debuts at #1 on DeepSWE. It outscores the previous best by 3% and sets a new state-of-the-art on our long-horizon coding benchmark.
108
84
55
1,617
475,882
We've updated the Artificial Analysis Coding Agent Index, replacing SWE-Bench Pro with Datacurve's DeepSWE benchmark - the swap lifts Codex with GPT-5.5 (xhigh) above Claude Code with Opus 4.8 (max), while the newly released Claude Fable 5 (max) in Claude Code debuts at the top DeepSWE, built by @datacurve, writes its tasks from scratch rather than adapting them from public GitHub issues or pull requests, so no model has seen the solutions during training. That matters because SWE-Bench Pro, the benchmark it replaces in our Coding Agent Index, had grown gameable, with some models recovering the fix from the repository's commit history instead of solving the task. The swap reorders the index: Codex with GPT-5.5 (xhigh) rises from 65 to 76, overtaking Claude Code with Opus 4.8 (max) at 73. Claude Code with Fable 5 (max), which enters directly on the refreshed index, leads at 77. SWE-Bench Pro had been flattering some combinations and penalizing others. More below.
111
177
104
1,944
590,163
WEEEEEEEEEEE
Datacurve
@datacurveMay 30
Opus 4.8 is now on DeepSWE. On the default high thinking effort, it scores 6% higher than Opus 4.7 xhigh, while also lowering average cost per task.
2
2
2
84
21,432
Leonard retweeted
Datacurve
@datacurveMay 30
Opus 4.8 is now on DeepSWE. On the default high thinking effort, it scores 6% higher than Opus 4.7 xhigh, while also lowering average cost per task.
87
116
107
1,750
987,144
Leonard retweeted
Neesh 🥭
@Neesh774May 27
The new @datacurve site is insane
11
6
3
509
29,267
Leonard retweeted
Theo - t3.gg
@theoMay 26
This is the first code bench that actually aligns with how it feels to use these models coding.
Serena Ge
@serenaa_geMay 26
Today we’re releasing DeepSWE, a new standard for agentic coding benchmarks. On public leaderboards, top models often look relatively close in capability. DeepSWE shows where they actually diverge, reflecting the realistic experience of developers in their day-to-day work.
118
149
10
3,596
306,107
Worked on acquiring and processing these tasks! Hope this provides unique, and valuable insights on how to better evaluate models in a sea of benchmarks
Serena Ge
@serenaa_geMay 26
Today we’re releasing DeepSWE, a new standard for agentic coding benchmarks. On public leaderboards, top models often look relatively close in capability. DeepSWE shows where they actually diverge, reflecting the realistic experience of developers in their day-to-day work.
2
16
19,727
Today we’re announcing we’ve raised $17.5 million in funding across a $15M Series A led by Chemistry and a $2.7M Seed to accelerate foundation model progress through providing frontier training data for LLMs. When we first started Datacurve, it came from a simple realization: foundation model progress is limited not just by compute, but by data quality and complexity. The right data unlocks new capabilities, especially in coding, where accuracy and reasoning matter most. We’re now proud to partner with the world’s leading foundation-model labs, providing them with high-quality, complex training data that helps push the boundaries of what AI can do. This is still just the start. Come build the future of technology with us in San Francisco: datacurve.ai/careers Huge thanks to our incredible team and investors who’ve believed in us since day one and beyond: @garrytan at @ycombinator, @1vnzh from @cohere , @Mark_Goldberg_ from @chemistry_fund, @TheDerrickLi from @AforeVC, @forwarddeploy, @SoheilK, and @shyamalanadkat.
149
60
23
1,056
450,493
Leonard retweeted
caps@capsjpeg22 May 2025
think i need more degods
12
11
103
5,375
launched v1 of Profiles. get in discord and let us know what features to add next :)
30
49
6
223
22,066
Leonard retweeted
we5lie@We5lie22 May 2025
DeGod Profiles V1 is live. Designed alongside with @budaz__ — excited to share what we’ve built for the community.
42
55
8
228
18,667
Leonard retweeted
adil
@adilcreates22 May 2025
Profiles v1 is live. Designed by @We5lie & @budaz__ — animated by me. Massive props to @LeonardMainnet, @0x_chill & @thebasedbob for engineering magic. Grateful to help bring this to life for the community.
18
15
97
4,832
Topped it up to 50 DeGods Fuck it, we ball @DeGodsNFT
BeByDay
@BeAlterEgos12 May 2025
Swept 20 DeGods Art slaps, always will. And Pasta cracks me up. @DeGodsNFT
65
39
2
242
11,907
I actually dropped a y00ts coloring book on Amazon recently! Please help spread the word 🙏 a.co/d/axYbcJD
130
158
10
660
131,581
It has always been @DeGodsNFT.
6
7
1
94
2,622
Load more