@estebankiwiKimi K2.7 Code is outperforming its predecessor across the board while using far fewer tokens. On Program Bench, it improved from 48% to 53% while cutting token consumption from 176k to 102k per task , a 42% reduction with a higher score. On Kimi Code Bench v2, it jumped from 51% to 62% while using 23% fewer tokens. The full details are in the thread: https://x.com/bridgemindai/status/2065430937861525553
Originalbeitrag ansehenFoto von @estebankiwi· Jun 12, 2026· Tech
The image shows a graph comparing the performance of Kimi-K2.7 Code and Kimi-K2.6. The graph is titled "Kimi-K2.7 Code vs Kimi-K2.6: Performance vs Tokens". The focus is on a line graph with data points for Kimi Code Bench v2, Program Bench, and MLS Bench Lite. The graph shows performance on the y-axis and tokens on the x-axis, with arrows indicating improvement from Kimi-K2.6 to Kimi-K2.7 Code for two of the data points. The mood/style is technical and analytical. Notable visual elements include a legend in the top-right corner distinguishing between Kimi-K2.6 and Kimi-K2.7 Code with gray and blue colors. There is no on-screen text beyond the graph's labels and title.
2Autonomous plane US flight milestone
Life of Developers infographic
SuperiorTrade Hyperliquid terminal
ChatGPT email screenshot
Hulk meme format
Vibe Coding vs Vibe Debugging meme
Will Smith futuristic gun robots
2Travis Scott RHYNO Spotify debut
20 JEV Tips for Machine Decision Models
Elon Musk Uranus joke theory
thv collage
Nick Bostrom on machine intelligence
Elon Musk X usage records
Audio app beats ChatGPT to #2
Apple executive on iPhone durability
Secret internet life and Pocket FM milestone
Google Gemini security breach alert
PS4 console update@estebankiwiKimi K2.7 Code is outperforming its predecessor across the board while using far fewer tokens. On Program Bench, it improved from 48% to 53% while cutting token consumption from 176k to 102k per task , a 42% reduction with a higher score. On Kimi Code Bench v2, it jumped from 51% to 62% while using 23% fewer tokens. The full details are in the thread: https://x.com/bridgemindai/status/2065430937861525553
Originalbeitrag ansehenFoto von @estebankiwi· Jun 12, 2026· Tech
The image shows a graph comparing the performance of Kimi-K2.7 Code and Kimi-K2.6. The graph is titled "Kimi-K2.7 Code vs Kimi-K2.6: Performance vs Tokens". The focus is on a line graph with data points for Kimi Code Bench v2, Program Bench, and MLS Bench Lite. The graph shows performance on the y-axis and tokens on the x-axis, with arrows indicating improvement from Kimi-K2.6 to Kimi-K2.7 Code for two of the data points. The mood/style is technical and analytical. Notable visual elements include a legend in the top-right corner distinguishing between Kimi-K2.6 and Kimi-K2.7 Code with gray and blue colors. There is no on-screen text beyond the graph's labels and title.
2Autonomous plane US flight milestoneLife of Developers infographicSuperiorTrade Hyperliquid terminalChatGPT email screenshotHulk meme formatVibe Coding vs Vibe Debugging memeWill Smith futuristic gun robots2Travis Scott RHYNO Spotify debut20 JEV Tips for Machine Decision ModelsElon Musk Uranus joke theorythv collageNick Bostrom on machine intelligenceElon Musk X usage recordsAudio app beats ChatGPT to #2Apple executive on iPhone durabilitySecret internet life and Pocket FM milestoneGoogle Gemini security breach alertPS4 console update