Ramp Labs's banner

Ramp Labs

@RampLabs • 14,969 subscribers

Shorts

We partnered with Prime Intellect to build Fast Ask, a small RL-trained subagent that helps our Sheets agent find answers in spreadsheets. It scores +4% over Opus on exact match accuracy at Haiku latency.

We partnered with Prime Intellect to build Fast Ask, a small RL-trained subagent that helps our Sheets agent find answers in spreadsheets. It scores +4% over Opus on exact match accuracy at Haiku latency.

334,154 次观看

Your work is now saved in Ramp Sheets. Every sheet you build automatically backs up to the cloud. Open the menu to jump back into any past project instantly.

Your work is now saved in Ramp Sheets. Every sheet you build automatically backs up to the cloud. Open the menu to jump back into any past project instantly.

17,662 次观看

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

Today, we’re launching Ramp Sheets, the AI spreadsheet editor built to 10x finance teams, operators, and founders. Ramp Sheets plans, executes, and formats automatically. It’s already being used to model budgets, forecast revenue, and reconcile financial statements.

Today, we’re launching Ramp Sheets, the AI spreadsheet editor built to 10x finance teams, operators, and founders. Ramp Sheets plans, executes, and formats automatically. It’s already being used to model budgets, forecast revenue, and reconcile financial statements.

1,327,472 次观看 • 11 个月前

Today, we're launching Ramp Agent Cards. There's been no safe way for agents to spend money, until now. Ramp Agent Cards give agents the ability to spend, governed with real spend limits, merchant controls, and full visibility into every transaction.

Today, we're launching Ramp Agent Cards. There's been no safe way for agents to spend money, until now. Ramp Agent Cards give agents the ability to spend, governed with real spend limits, merchant controls, and full visibility into every transaction.

849,245 次观看 • 7 个月前

Today, we're releasing Ramp CLI to let agents manage your company's finances. 50+ tools across cards, bills, expenses, travel, and approvals. Fewer tokens than MCP, and comes with pre-built skills like receipt compliance and agentic purchasing.

Today, we're releasing Ramp CLI to let agents manage your company's finances. 50+ tools across cards, bills, expenses, travel, and approvals. Fewer tokens than MCP, and comes with pre-built skills like receipt compliance and agentic purchasing.

604,733 次观看 • 7 个月前

Introducing Ramp Accounting Bench. We partnered with accounting professionals to create 137 tasks and grading rubrics grounded in everyday accounting workflows. Even with three attempts, the best model achieved only 21% accuracy. Reliable agentic accounting remains an open challenge.

Introducing Ramp Accounting Bench. We partnered with accounting professionals to create 137 tasks and grading rubrics grounded in everyday accounting workflows. Even with three attempts, the best model achieved only 21% accuracy. Reliable agentic accounting remains an open challenge.

88,328 次观看 • 1 个月前

Meta just acquired Manus AI. Ramp Sheets modeled it out: Estimated price: $4-6B based on AI M&A comps Fastest to $100M ARR in history (8 months) Benchmark likely 8-12x'd in under a year

Meta just acquired Manus AI. Ramp Sheets modeled it out: Estimated price: $4-6B based on AI M&A comps Fastest to $100M ARR in history (8 months) Benchmark likely 8-12x'd in under a year

528,139 次观看 • 10 个月前

We’re open-sourcing PorTAL, our framework for shared task representations and cross model LoRA adaptation. It now spans from hybrid attention models to multimodal systems including Gemma 4 E2B, Mistral 7B & Thinking Machines' Inkling. Code: ramp-public/portallib Models: Hugging Face /RampPublic

We’re open-sourcing PorTAL, our framework for shared task representations and cross model LoRA adaptation. It now spans from hybrid attention models to multimodal systems including Gemma 4 E2B, Mistral 7B & Thinking Machines' Inkling. Code: ramp-public/portallib Models: Hugging Face /RampPublic

171,845 次观看 • 3 个月前

Introducing Steer AI. We made an AI that can't stop thinking about any concept you choose, by steering a model's internal representations at inference time. Ask it anything, and watch it bend reality around that concept. Available for one week only.

Introducing Steer AI. We made an AI that can't stop thinking about any concept you choose, by steering a model's internal representations at inference time. Ask it anything, and watch it bend reality around that concept. Available for one week only.

348,570 次观看 • 6 个月前

Introducing PorTAL: Portable Task Adapters for LLMs. A novel recipe to cheaply port fine-tuning between models. It matches per task LoRA accuracy at half the cost, lowering the switching overhead of adapting tasks across LLMs. At Ramp, every new model release used to mean retraining our fine-tunes from scratch. PorTAL learns the task once, then efficiently refits it onto any new base model, even across model families.

Introducing PorTAL: Portable Task Adapters for LLMs. A novel recipe to cheaply port fine-tuning between models. It matches per task LoRA accuracy at half the cost, lowering the switching overhead of adapting tasks across LLMs. At Ramp, every new model release used to mean retraining our fine-tunes from scratch. PorTAL learns the task once, then efficiently refits it onto any new base model, even across model families.

173,290 次观看 • 3 个月前

We tested NVIDIA NeMo Switchyard’s stage router for coding agents in Ramp SWE-Bench. Routed agents showed comparable performance to single-model controls while substantially reducing costs and runtime.

We tested NVIDIA NeMo Switchyard’s stage router for coding agents in Ramp SWE-Bench. Routed agents showed comparable performance to single-model controls while substantially reducing costs and runtime.

100,276 次观看 • 2 个月前

We modeled the SpaceX / xAI merger. – xAI holders get 22% of combined entity – $49B revenue by 2028 – SpaceX currently trades at 50x revenue At that multiple, the combined entity is worth ~$1.1T at IPO and $2.5T by 2028.

We modeled the SpaceX / xAI merger. – xAI holders get 22% of combined entity – $49B revenue by 2028 – SpaceX currently trades at 50x revenue At that multiple, the combined entity is worth ~$1.1T at IPO and $2.5T by 2028.

275,730 次观看 • 9 个月前

We had early access to NVIDIA AI Nemotron 3.5 Super VL and post-trained it for one of our highest-volume workflows: receipt extraction. One epoch of supervised fine-tuning on 10,000 examples made it competitive with leading open and closed-source models on our receipt benchmark.

We had early access to NVIDIA AI Nemotron 3.5 Super VL and post-trained it for one of our highest-volume workflows: receipt extraction. One epoch of supervised fine-tuning on 10,000 examples made it competitive with leading open and closed-source models on our receipt benchmark.

16,168 次观看 • 12 天前

xAI just raised $20B at ~$230B. Ramp Sheets thinks it's a good deal: $383/user vs OpenAI's $1,875/user (80% discount) 600M users who already open X daily Only need 8% converting to $20/mo Grok to justify valuation

xAI just raised $20B at ~$230B. Ramp Sheets thinks it's a good deal: $383/user vs OpenAI's $1,875/user (80% discount) 600M users who already open X daily Only need 8% converting to $20/mo Grok to justify valuation

249,299 次观看 • 9 个月前

AI spend is measured in tokens, model calls, and dollars. However, none of these fields describe the actual work being done. At Ramp, we built a semantic layer that attributes agent spend to objectives and outcomes. This allowed us to go from monitoring AI spend to understanding AI ROI. Here's how we built it 🧵

AI spend is measured in tokens, model calls, and dollars. However, none of these fields describe the actual work being done. At Ramp, we built a semantic layer that attributes agent spend to objectives and outcomes. This allowed us to go from monitoring AI spend to understanding AI ROI. Here's how we built it 🧵

44,997 次观看 • 1 个月前

We built Anthropic's internal revenue model based on the latest reports. $4.5B → $148B in 5 years $100B in training costs Profitable two years before OpenAI

We built Anthropic's internal revenue model based on the latest reports. $4.5B → $148B in 5 years $100B in training costs Profitable two years before OpenAI

147,905 次观看 • 9 个月前

Recent KV cache compression research has focused on small models with dense attention. We scaled these techniques to GLM 5.3, a frontier sparse attention model. We adapted KVzip and Attention Matching, reducing KV cache size by 80% while retaining over 90% of full-context accuracy on the QuALITY benchmark.

Recent KV cache compression research has focused on small models with dense attention. We scaled these techniques to GLM 5.3, a frontier sparse attention model. We adapted KVzip and Attention Matching, reducing KV cache size by 80% while retaining over 90% of full-context accuracy on the QuALITY benchmark.

22,064 次观看 • 29 天前

A finance team just sent us this. They ran a full budget vs. actuals analysis in under 5 minutes. Uploaded their budget P&L and 10 months of GL actuals. Ramp Sheets matched every line item, calculated variances, and flagged problem areas. This used to take half a day.

A finance team just sent us this. They ran a full budget vs. actuals analysis in under 5 minutes. Uploaded their budget P&L and 10 months of GL actuals. Ramp Sheets matched every line item, calculated variances, and flagged problem areas. This used to take half a day.

141,898 次观看 • 9 个月前

B2B SaaS switching costs collapse during competitor acquisitions. – 15% of their customer base enters consideration – CAC drops 3-5x – Conversion spikes 5-10x Capture a few % = years of compounding LTV. Plug in your competitor's customer count and see the opportunity ↓

B2B SaaS switching costs collapse during competitor acquisitions. – 15% of their customer base enters consideration – CAC drops 3-5x – Conversion spikes 5-10x Capture a few % = years of compounding LTV. Plug in your competitor's customer count and see the opportunity ↓

138,780 次观看 • 9 个月前

We deployed 10,000 background agents to security-scan our codebase. The system is simple, scales with compute, and runs on publicly available models. From the scan, we fixed several high-severity vulnerabilities.

We deployed 10,000 background agents to security-scan our codebase. The system is simple, scales with compute, and runs on publicly available models. From the scan, we fixed several high-severity vulnerabilities.

75,562 次观看 • 5 个月前

Anthropic just raised $30B at a $380B valuation. – $0 to $14B in revenue in under 3 years – 95x return for Series A investors – If they IPO at $1T, new investors 3x their money View the full investment analysis and path to $1T below.

Anthropic just raised $30B at a $380B valuation. – $0 to $14B in revenue in under 3 years – 95x return for Series A investors – If they IPO at $1T, new investors 3x their money View the full investment analysis and path to $1T below.

112,219 次观看 • 8 个月前

You can now upload files to Ramp Sheets. Watch the agent ingest a pitch deck, research competitors on the web, and reverse-engineer a full Operating Model & Valuation from a single PDF.

You can now upload files to Ramp Sheets. Watch the agent ingest a pitch deck, research competitors on the web, and reverse-engineer a full Operating Model & Valuation from a single PDF.

110,119 次观看 • 10 个月前