# Weave vs Faros

Canonical: https://weaveos.com/compare/weave-vs-faros

Title: Weave vs Faros: code effort, agent outcomes and AI costs

Weave estimates the expert effort represented by code changes and explains the intent, complexity, risk and dependencies behind each score. It combines that measure with AI cost, session context, reviews and rework. Faros uses customer-defined successful outcomes for its published cost-per-outcome method; Weave measures the work inside the code change. Weave reports average token-spend savings of 40% with its separately offered Router, and 20–30% faster shipping for customers using Weave.

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Weave AI ROI methodology: output units and cost calculations](https://weaveos.com/methodology/ai-roi); [Faros cost per outcome: definition and examples](https://www.faros.ai/blog/cost-per-outcome); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform); [Faros product overview](https://www.faros.ai/); [Weave Router: separately offered coding-agent routing](https://weaveos.com/products/router); [Weave’s July 2026 reported product results](https://weaveos.com/blog/weave-raises-13.5m-series-a-to-build-engineering-and-token-intelligence)

[Book a demo](https://weaveos.com/demo) · [Get started for free](https://app.weaveos.com/sign-up)

Sources reviewed: 2026-11-13

## The short version

Estimated code effort within each successful outcome.

### Weave

Effort inside completed outcomes

- Expert-effort estimates for the resulting code
- AI output per million tokens
- Bug-introduction traces from later merged fixes

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Weave benchmarks: cost, output per million tokens and peer populations](https://docs.weaveos.com/annotations/benchmarks); [Bugs introduced: source attribution, confidence and timing](https://docs.weaveos.com/metrics/bugs-introduced)

#### What is Weave?

Weave measures the engineering effort inside code that an agent helped deliver. A successful routine update and a successful migration can represent different amounts of work; Silk 1 explains that difference and supports calibration. Cost per estimated expert hour, output per million tokens, review analysis and bug-introduction traces give a team separate ways to examine the result behind a completed task.

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Token Intelligence: AI usage, cost sources and attribution](https://weaveos.com/products/token-intelligence); [Engineering Intelligence: quality, delivery and developer feedback](https://weaveos.com/products/engineering-intelligence); [Weave Agent Observability: connected session evidence](https://weaveos.com/agent-observability); [AI Skills V2: human decisions, eligible sessions and assessment evidence](https://docs.weaveos.com/releases/ai-skills-v2); [Wooly: answers grounded in engineering records](https://weaveos.com/products/wooly); [Weave AI ROI report: financial value, costs and rate inputs](https://docs.weaveos.com/reports/ai-roi); [Weave custom dashboards: authoring and sharing](https://docs.weaveos.com/features/custom-dashboards)

### Faros

Engineering graph and cost attribution

- Sessions and engineering records connected in a graph
- Spend attribution by session, PR, team and model
- Route evaluation and policy enforcement

Sources: [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform); [Faros product overview](https://www.faros.ai/)

#### What is Faros?

Faros, formerly Faros AI, builds an Engineering World Model connecting sessions, tickets, commits, PRs and CI outcomes. Its platform includes attribution, evaluation of model routes on completed work, and policy controls for budgets, models and routing.

Sources: [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform); [Faros product overview](https://www.faros.ai/)

## How Weave and Faros compare

| What matters | Weave | Faros |
| --- | --- | --- |
| What counts as output | Silk 1 estimates expert effort from code intent, complexity, risk and cross-file dependencies. Each score has written reasoning and supports team calibration. | Cost per verified successful outcome defined in advance, using direct vendor costs or fully loaded costs that include estimated human review and correction. |
| What the AI budget produced | See how AI spending compares with the effort in the code shipped by the same team during that period. Cost labels tell you which figures come from bills, usage estimates or subscriptions. | Spend attribution by session, PR, team and model. |
| AI contribution and sessions | See AI usage and authorship alongside code output. Open captured agent sessions to understand how the work happened; detail depends on the connected tool and setup. | Sessions linked to engineering records and outcomes. |
| Quality and delivery | Review feedback scored for depth, practicality and impact, alongside cycles, reverts, turnover and developer feedback. | Code and CI evidence in an engineering graph. |
| Model routing and controls | Weave Router is a separately offered open-source coding-agent router with cache-aware selection and escalation. | Route evaluation, approved routes, budgets and policy enforcement. |

**What counts as output**

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform); [Faros cost per outcome: definition and examples](https://www.faros.ai/blog/cost-per-outcome)

**What the AI budget produced**

Sources: [Token Intelligence: AI usage, cost sources and attribution](https://weaveos.com/products/token-intelligence); [Weave AI ROI methodology: output units and cost calculations](https://weaveos.com/methodology/ai-roi); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform)

**AI contribution and sessions**

Sources: [Token Intelligence: AI usage, cost sources and attribution](https://weaveos.com/products/token-intelligence); [Weave Agent Observability: connected session evidence](https://weaveos.com/agent-observability); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform)

**Quality and delivery**

Sources: [Engineering Intelligence: quality, delivery and developer feedback](https://weaveos.com/products/engineering-intelligence); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform); [Weave review quality: depth, communication, practicality and impact](https://docs.weaveos.com/metrics/review-quality)

**Model routing and controls**

Sources: [Weave Router: separately offered coding-agent routing](https://weaveos.com/products/router); [Weave Router: public source and installation](https://github.com/weave-os/router); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform); [Faros product overview](https://www.faros.ai/)

## Code-output measurement

### Weave: estimated expert effort

Weave measures estimated expert effort inside the code change. Two successful tasks can therefore represent different amounts of engineering work. Engineers can inspect the score’s reasoning and calibrate the estimate before comparing it with AI spending.

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Weave AI ROI methodology: output units and cost calculations](https://weaveos.com/methodology/ai-roi); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform); [Faros product overview](https://www.faros.ai/)

### Faros’s approach

Faros links costs to customer-defined successful outcomes and can include human review and correction costs.

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Weave AI ROI methodology: output units and cost calculations](https://weaveos.com/methodology/ai-roi); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform); [Faros product overview](https://www.faros.ai/)

Why Weave

## What Weave adds over Faros

### Expert-effort output

Faros links cost to verified outcomes. Weave adds estimated expert effort for the code within those outcomes: a successful routine update can represent less work than a successful migration. Silk 1 explains the estimate and accepts team corrections. The effort unit allows the code contribution to change even when the number of completed tasks does not. In Weave’s published evaluation, Silk 1 had 3.3× lower average prediction error than the previous Weave model; 88% of PR scores were within ±1 hour of an expert estimate.

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Weave AI ROI methodology: output units and cost calculations](https://weaveos.com/methodology/ai-roi)

### Output per token

Faros’s successful-outcome denominator establishes whether a task reached its goal. Weave’s output per million tokens examines how much AI-attributed expert effort accompanied the token consumption. It includes code-generating tools with both output and token telemetry for the period. Compare that ratio with cost per output, whose numerator also reflects prices and seat costs. Inspect the underlying code and review trends to understand an increase in consumption rather than assume that completing more tasks required the same amount of work.

Sources: [Token Intelligence: AI usage, cost sources and attribution](https://weaveos.com/products/token-intelligence); [Weave AI tool costs: billing, estimates and subscriptions](https://docs.weaveos.com/metrics/ai-tool-costs); [Weave AI ROI methodology: output units and cost calculations](https://weaveos.com/methodology/ai-roi); [Weave benchmarks: cost, output per million tokens and peer populations](https://docs.weaveos.com/annotations/benchmarks); [Weave AI ROI report: financial value, costs and rate inputs](https://docs.weaveos.com/reports/ai-roi)

### Bugs traced to their source

A completed task can still introduce a bug found later. When a bug-fix PR merges, Weave traces modified or deleted lines through Git history, evaluates candidate source commits and records confidence in the likely introduction. Medium, high and very-high confidence attributions enter the metric; speculative low-confidence candidates do not. The bug is attributed to when its source code was written, so an older period can change as fixes arrive. Read these traces alongside the code-output and review trends.

Sources: [Engineering Intelligence: quality, delivery and developer feedback](https://weaveos.com/products/engineering-intelligence); [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Weave review quality: depth, communication, practicality and impact](https://docs.weaveos.com/metrics/review-quality); [Bugs introduced: source attribution, confidence and timing](https://docs.weaveos.com/metrics/bugs-introduced)

### Human AI Skills

Weave can connect a completed code change with a discussion of how the engineer directed the agent. AI Skills V2 assesses the person’s decisions and explains the rating at a specific conversation location. The manager can review a planning or verification example rather than infer skill from the outcome alone. Daily sampling selects one qualifying session per engineer; a weekly fallback covers engineers with no daily pick. Skills with no relevant opportunity remain unscored.

Sources: [AI Skills V2: human decisions, eligible sessions and assessment evidence](https://docs.weaveos.com/releases/ai-skills-v2); [Engineering Intelligence: quality, delivery and developer feedback](https://weaveos.com/products/engineering-intelligence)

### Wooly and MCP

When a completed outcome contains substantial code work, Wooly can investigate the effort estimate and link its answer to the underlying records. A compatible MCP client can retrieve available metrics, chart drilldowns, PRs and reviews using the connected user’s permissions. Configured project categories provide work context. Both paths support examining a team-level result without exporting and manually reconnecting separate charts; the MCP tools are read-only.

Sources: [Wooly: answers grounded in engineering records](https://weaveos.com/products/wooly); [Weave MCP: read-only metrics and engineering records](https://docs.weaveos.com/integrations/mcp); [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence)

### Financial AI ROI

Cost per successful outcome and financial return answer different questions. Weave estimates Value from AI using either additional code output above a size-cohort pre-AI benchmark, capped by AI-attributed output, or the AI share of engineering spend. It then deducts AI tool cost to show Net savings and ROI. The Output formula retains the distinction between a routine completed task and a larger code change by using estimated expert effort. The report exposes the formula and engineering-cost inputs for review.

Sources: [Weave AI ROI report: financial value, costs and rate inputs](https://docs.weaveos.com/reports/ai-roi); [Weave AI ROI methodology: output units and cost calculations](https://weaveos.com/methodology/ai-roi); [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence)

### Open-source model routing

Weave Router publishes its implementation and setup for supported coding clients, with an Apache-2.0 self-hosting path. Weave reports average token-spend savings of 40% with Router; its September 2026 paired trials used 46–52% of direct GPT-6 Astra’s cost per trial. Faros also evaluates and enforces routes. Choose Weave’s separate router when source access, client setup and the published coding evaluation match how you want to run the service.

Sources: [Weave Router: separately offered coding-agent routing](https://weaveos.com/products/router); [Weave Router: public source and installation](https://github.com/weave-os/router); [Weave Router 2.0: September 2026 paired coding benchmarks](https://weaveos.com/blog/introducing-weave-router-2-0); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform)

## Using Weave and Faros

### Complex code changes

**Weave:** Read Silk 1’s expert-effort estimate and reasoning. Make complexity visible before comparing the work with AI spending.

**Faros:** Sessions linked to engineering work.

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform)

### Token efficiency

**Weave:** If token use rises faster than output, open the work before changing the budget. Compare AI-attributed output per million tokens with cost and quality, then save the review in a dashboard. The ratio includes tools with both output and token data and shows the peer group used.

**Faros:** Spend by session, PR, team and model.

Sources: [Weave benchmarks: cost, output per million tokens and peer populations](https://docs.weaveos.com/annotations/benchmarks); [Weave custom dashboards: authoring and sharing](https://docs.weaveos.com/features/custom-dashboards); [Weave custom queries: plain English, SQL, preview and drilldown](https://docs.weaveos.com/features/custom-queries); [Weave AI ROI methodology: output units and cost calculations](https://weaveos.com/methodology/ai-roi); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform)

### Model routing

**Weave:** Use the separately offered Weave Router to test cache-aware model selection and escalation inside supported coding agents.

**Faros:** Route evaluation and model/budget policies.

Sources: [Weave Router: separately offered coding-agent routing](https://weaveos.com/products/router); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform)

## When to choose Weave

Choose Weave when you need the code inside an outcome measured in estimated expert effort, then connected to costs, reviews and human AI skills. That makes a changing mix of engineering work easier to investigate and compare.

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Engineering Intelligence: quality, delivery and developer feedback](https://weaveos.com/products/engineering-intelligence); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform); [Faros product overview](https://www.faros.ai/)

## Frequently asked questions

### Which AI cost-per-output metric accounts for the work inside each change?

Weave estimates the expert effort represented by a code change. The score reflects intent, complexity, risk and dependencies, with written reasoning and team calibration. That distinguishes the work inside a routine update from the work inside a difficult migration before comparing AI costs.

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Weave AI ROI methodology: output units and cost calculations](https://weaveos.com/methodology/ai-roi); [Faros cost per outcome: definition and examples](https://www.faros.ai/blog/cost-per-outcome)

### Does Faros link sessions to PRs and AI costs?

Yes. Faros describes session and PR attribution within its engineering graph. Weave’s distinction here is its explained expert-effort output model.

Sources: [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform); [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence)

### Is model routing unique to Weave?

No. Faros evaluates model routes and enforces policies through its own or an external router. Weave Router is a separately offered open-source coding-agent router, with its own supported clients, cache behavior and escalation controls.

Sources: [Faros product overview](https://www.faros.ai/); [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform); [Weave Router: separately offered coding-agent routing](https://weaveos.com/products/router); [Weave Router: public source and installation](https://github.com/weave-os/router)

### Does cost per output include every AI tool?

The ratio matches the cost of code-generating tools with AI-attributed code output for the same period. Review-only tools are excluded from that code-output calculation. Their spending and review contribution require a separate review analysis.

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Weave AI ROI methodology: output units and cost calculations](https://weaveos.com/methodology/ai-roi); [Weave AI ROI report: financial value, costs and rate inputs](https://docs.weaveos.com/reports/ai-roi)

### How does Weave turn an AI trend into a coaching discussion?

AI Skills V2 assesses the engineer’s decisions in a sample of captured conversations that meet its scoring requirements. Open the rating, explanation and the part of the conversation behind it to discuss planning, verification or context. A skill is left unscored when the conversation gave no opportunity to demonstrate it. Use the conversation example for coaching, then track output and quality separately.

Sources: [AI Skills V2: human decisions, eligible sessions and assessment evidence](https://docs.weaveos.com/releases/ai-skills-v2); [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Engineering Intelligence: quality, delivery and developer feedback](https://weaveos.com/products/engineering-intelligence)

### Can we investigate the work from Claude or Cursor?

Yes. Weave’s read-only MCP server lets compatible clients query the engineering metrics available through MCP and inspect the PRs, reviews and work grouped by configured project categories behind them with the same access permissions as the user. That gives an engineering leader another way to investigate the explained output behind a cost trend.

Sources: [Weave MCP: read-only metrics and engineering records](https://docs.weaveos.com/integrations/mcp); [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence)

### How can engineers review an output score?

Inspect the code and written effort estimate for the change behind the completed outcome. A larger migration and a routine update can both be completed tasks yet represent different engineering effort. Teams can submit corrections to the estimate and use the calibrated result alongside delivery, review and later bug-introduction traces.

Sources: [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence); [Engineering Intelligence: quality, delivery and developer feedback](https://weaveos.com/products/engineering-intelligence)

## See Weave on your repositories

Connect your repositories and AI tools to see code-output scores, AI costs, reviews and delivery for your team. Engineers can inspect the reasoning for individual PRs and correct estimates with missing context.

[Sign up](https://app.weaveos.com/sign-up) · [Book a demo](https://weaveos.com/demo)

## Sources and comparison method

Written by Weave using the linked first-party product pages and documentation, checked November 13, 2026. This is a comparison of products and measurement methods, not a head-to-head performance test. Packaging and integration coverage can change.

- [Weave Router 2.0: September 2026 paired coding benchmarks](https://weaveos.com/blog/introducing-weave-router-2-0)
- [Weave AI ROI report: financial value, costs and rate inputs](https://docs.weaveos.com/reports/ai-roi)
- [Silk 1: explained code output and team calibration](https://weaveos.com/products/code-intelligence)
- [Token Intelligence: AI usage, cost sources and attribution](https://weaveos.com/products/token-intelligence)
- [Weave AI ROI methodology: output units and cost calculations](https://weaveos.com/methodology/ai-roi)
- [Engineering Intelligence: quality, delivery and developer feedback](https://weaveos.com/products/engineering-intelligence)
- [Weave Agent Observability: connected session evidence](https://weaveos.com/agent-observability)
- [Wooly: answers grounded in engineering records](https://weaveos.com/products/wooly)
- [Weave Router: separately offered coding-agent routing](https://weaveos.com/products/router)
- [Faros product overview](https://www.faros.ai/)
- [Faros platform: work graph, attribution, and routing evaluation](https://www.faros.ai/platform)
- [Weave Router: public source and installation](https://github.com/weave-os/router)
- [Faros cost per outcome: definition and examples](https://www.faros.ai/blog/cost-per-outcome)
- [AI Skills V2: human decisions, eligible sessions and assessment evidence](https://docs.weaveos.com/releases/ai-skills-v2)
- [Weave custom dashboards: authoring and sharing](https://docs.weaveos.com/features/custom-dashboards)
- [Weave custom queries: plain English, SQL, preview and drilldown](https://docs.weaveos.com/features/custom-queries)
- [Weave benchmarks: cost, output per million tokens and peer populations](https://docs.weaveos.com/annotations/benchmarks)
- [Weave MCP: read-only metrics and engineering records](https://docs.weaveos.com/integrations/mcp)
- [Weave AI tool costs: billing, estimates and subscriptions](https://docs.weaveos.com/metrics/ai-tool-costs)
- [Weave’s July 2026 reported product results](https://weaveos.com/blog/weave-raises-13.5m-series-a-to-build-engineering-and-token-intelligence)
- [Weave review quality: depth, communication, practicality and impact](https://docs.weaveos.com/metrics/review-quality)
- [Bugs introduced: source attribution, confidence and timing](https://docs.weaveos.com/metrics/bugs-introduced)

## Inspect the evidence

- [How Weave calculates AI cost per unit of output](https://weaveos.com/methodology/ai-roi)
- [Inspect the output model and score reasoning](https://weaveos.com/products/code-intelligence)
- [Weave benchmarks: cost, output per million tokens and peer populations](https://docs.weaveos.com/annotations/benchmarks)
- [AI Skills V2: human decisions, eligible sessions and assessment evidence](https://docs.weaveos.com/releases/ai-skills-v2)
