| --- |
| license: mit |
| language: |
| - en |
| base_model: |
| - LiquidAI/LFM2.5-230M |
| --- |
| ## LFM 2.5 230m Code-Enhanced |
| Lfm 2.5 230m code enhanced is a code-ish version of the previous lfm 2.5 230m (base) , this model however attempts to make the old model better at coding |
| this uses the latest weight space programming paper along with deepseek v4.1 flash to manually change the model weights. |
|
|
| <!DOCTYPE html> |
| <html lang="en"> |
| <head> |
| <meta charset="UTF-8"> |
| <meta name="viewport" content="width=device-width, initial-scale=1"> |
| <title>LFM CodeEnhanced AI</title> |
| <style> |
| * { box-sizing: border-box; } |
| body { |
| margin: 0; |
| min-height: 100vh; |
| display: grid; |
| place-items: center; |
| background: #080b14; |
| font-family: Inter, Arial, sans-serif; |
| } |
| .logo { |
| display: flex; |
| align-items: center; |
| gap: 18px; |
| } |
| svg { width: 100px; height: 100px; } |
| .brand-name { |
| font-size: 30px; |
| font-weight: 850; |
| letter-spacing: -1.5px; |
| color: black; |
| } |
| .brand-name span { |
| background: linear-gradient(100deg,#00e5ff,#9361ff); |
| -webkit-background-clip: text; |
| background-clip: text; |
| color: transparent; |
| } |
| .tagline { |
| margin-top: 7px; |
| color: #8996b5; |
| font-size: 10px; |
| font-weight: 700; |
| letter-spacing: 4px; |
| } |
| </style> |
| </head> |
| <body> |
| <div class="logo"> |
| <svg viewBox="0 0 100 100" fill="none"> |
| <defs> |
| <linearGradient id="g" x1="10" y1="10" x2="90" y2="90"> |
| <stop stop-color="#00e5ff"/> |
| <stop offset="1" stop-color="#9361ff"/> |
| </linearGradient> |
| </defs> |
| <path d="M50 4 91 27V73L50 96 9 73V27L50 4Z" |
| fill="#111a30" stroke="url(#g)" stroke-width="2.5"/> |
| <path d="m35 33-15 17 15 17M65 33l15 17-15 17" |
| stroke="url(#g)" stroke-width="5" |
| stroke-linecap="round" stroke-linejoin="round"/> |
| <path d="m50 23 6 21 21 6-21 6-6 21-6-21-21-6 21-6 6-21Z" |
| fill="url(#g)"/> |
| <circle cx="50" cy="50" r="5" fill="white"/> |
| </svg> |
| <div> |
| <div class="brand-name">LFM <span>CodeEnhanced</span></div> |
| <div class="tagline">AI · CODE · CREATE</div> |
| </div> |
| </div> |
| </body> |
| </html> |
| |
|
|
| Currently has 3 quantizations |
| -- |
| | Quantization | Quality | |
| |----------------|--------| |
| | BF16 | Highest quality <span style="background-color: #12CA12; color: black; padding: 2px 8px; border-radius: 10px; font-size: 12px; font-weight: bold;">Recommended</span> |
| | Q4_KM | Mid Quality | |
| | Q8 | Also good | |
| |
| --- |
| |
| This model was generated by deepseek V4.1 FLASH by changing model weights. |
| |
| All credits to <b>liquid-ai, deepseek & weight space programming research paper</b> |
| |
| --- |
