Sample report
Sample competitive landscape: AI coding assistants and agents
Sample report. Independently produced, not commissioned by any company named. Evidence captured 2026-10-08.
- Companies in file
- 8
- Distinct source URLs
- 21
- Claims by evidence grade
- A 53 · B 0 · C 4 · D 4
Executive conclusion
Recommendation
What this evidence supports: do not rely on a lower price or on new-model access as your wedge. Worth testing (hypotheses, not findings): workflow depth, vertical specialisation, enterprise controls.
Confidence
Confidence varies by finding — F1 Medium, F2 Low, F3 Medium — based on public pricing and launch pages only, eight vendors, evidence captured 2026-10-08.
F1
GRADE C · INTERPRETATIONConfidence: Medium — Based on public pricing pages for seven vendors with listed entry and/or power tiers, evidence captured 2026-10-08.
Entry-level paid individual tiers cluster around $10-$20 (Copilot Pro $10, JetBrains AI Pro $10, Cursor Pro $20, Devin Pro $20, Kiro Pro $20, Claude Pro $20, Amazon Q Pro $19), and most vendors also publish a power tier around $100-$200 (Copilot Max $100, Claude Max from $100, Cursor Ultra $200, Devin Max $200, Kiro Power $200). The price ladder is largely standardised, so a lower list price is unlikely to be a durable wedge.
- Evidence:
- GitHub Copilot, docs.github.com/en/copilot/get-started/plans — retrieved 2026-10-08 — Grade A
- JetBrains AI, jetbrains.com/help/ai-assistant/licensing-and-subscriptions.html — retrieved 2026-10-08 — Grade A
- Cursor, cursor.com/docs/models-and-pricing — retrieved 2026-10-08 — Grade A
- Devin, devin.ai/pricing — retrieved 2026-10-08 — Grade A
- Kiro, kiro.dev/pricing — retrieved 2026-10-08 — Grade A
- Claude Code, claude.com/pricing — retrieved 2026-10-08 — Grade A
- Amazon Q Developer, aws.amazon.com/q/developer/pricing — retrieved 2026-10-08 — Grade A
F2
GRADE C · INTERPRETATIONConfidence: Low — Only two launches in this capture, and both concern Claude 5.5 model availability.
Availability of a Claude 5.5 model was announced by two vendors within a week (Kiro: Sonnet 5.5, 2026-10-02; Devin: Haiku 5.5, 2026-10-07), which suggests new-model access is a fast-following feature rather than a lasting differentiator.
- Evidence:
- Devin, devin.ai/blog/claude-haiku-5-5 — retrieved 2026-10-08 — Grade A
- Kiro, kiro.dev/blog/sonnet-5-5 — retrieved 2026-10-08 — Grade A
F3
GRADE C · INTERPRETATIONConfidence: Medium — for the vendor-level statement that Amazon is moving its coding assistant onto Kiro, based on one Grade A official product-page notice.
Amazon is moving its coding assistant onto Kiro: Amazon Q Developer's page states IDE plugin support ends 2027-04-30 and points to Kiro.
This is evidence about one vendor's product line; it does not on its own show wider market consolidation.
- Evidence:
- Amazon Q Developer, aws.amazon.com/q/developer — retrieved 2026-10-08 — Grade A
What would change our mind
Conditions to monitor — not predictions.
- A vendor with an entry-level paid individual tier well outside the $10–$20 cluster gaining visible adoption on its own public packaging.
- A vendor publishing model-exclusive access as its headline packaging on an official product or pricing page.
- Amazon keeping Amazon Q Developer IDE plugins as an ongoing product past 2027-04-30, or withdrawing the official pointer to Kiro.
Scope and decision
Decision supported: Where a company building an AI coding product should and should not compete, based only on publicly stated positioning, packaging and recent launches of eight vendors.
Scope: Official product, pricing, documentation and changelog pages for up to eight vendors. No private interviews, no paid data sources, no invented figures.
Market: AI coding assistants and agents
Prices shown exactly as listed on each vendor's page. Currency is stated only where the vendor's page states it.
| Company | Target | Pricing | Free plan | Positioning | Launch |
|---|---|---|---|---|---|
| Cursor grades A/A/A/A/A/A | Developers and teams building software with AI coding agents (site frames Cursor as a coding agent for ambitious software; Teams and Enterprise plans address collaborating professionals and scaled organisations). | Hobby: Free · Pro: $20/mo · Pro Plus: $60/mo · Ultra: $200/mo · Teams Standard: $40/user/mo · Teams Premium: $120/user/mo · Enterprise: Custom · Start (India only): ₹649/mo, tax inclusive (INR) | Yes | Positions Cursor as a coding agent for building ambitious software, with agents that can plan and build, run in parallel, and work across desktop, CLI and connected tools. | Remote control for local agents |
| GitHub Copilot grades A/A/A/A/A/A | Individuals (Free, Student, Pro, Pro+, Max) and organisations or enterprises (Business, Enterprise), depending on plan. | Copilot Free: Free (USD) · Copilot Student: Free (USD) · Copilot Pro: $10 USD per month · Copilot Pro+: $39 USD per month · Copilot Max: $100 USD per month · Copilot Business: $19 USD per granted seat per month · Copilot Enterprise: $39 USD per granted seat per month | Yes | AI-powered coding assistance with plans spanning limited free access through individual and organisation/enterprise seats, including agents and a monthly GitHub AI Credits allowance on paid plans. | Copilot Max individual plan announced (with flex allotments for Pro and Pro+) |
| Claude Code grades A/A/A/A/A/D | Developers who want to hand Claude bug fixes, tests or multi-day migrations and steer work from terminal, IDE, Slack or web; access via Claude Pro, Max, Team, Enterprise or Claude Console. | Claude Free: $0 · Claude Pro: $20 if billed monthly; $17 per month with annual subscription discount ($200 billed up front) · Claude Max: From $100 per month · Claude Team Standard seat: $20 per seat / month if billed annually; $25 if billed monthly · Claude Team Premium seat: $100 per seat / month if billed annually; $125 if billed monthly · Claude Enterprise: $20 per seat per month plus usage billed at API rates (annual) | No | Agentic coding product that plans, asks clarifying questions and handles work that can run for hours or days, used from terminal, IDE, web and mobile. | Not verified in this capture |
| Devin grades A/A/A/A/A/A | Software teams seeking an autonomous software engineer that can run in the cloud or on a machine and deliver work through to a merge-ready pull request. | Free: $0 · Pro: $20 per month · Max: $200 per month · Teams: $80/month for team plan + $40/mo per full dev seat · Enterprise: Let's talk | Yes | Positions Devin as the team's autonomous software engineer that tests end to end on its own computers and keeps working until a pull request is ready to merge. | Claude Haiku 5.5 is now available in Devin |
| Amazon Q Developer grades A/A/A/A/A/D | Software developers building, securing, managing and optimising applications on or off AWS, including use in IDEs, CLI and AWS console contexts. | Free Tier: Advanced capabilities at zero cost (with stated monthly limits) · Pro Tier: $19/mo. per user | Yes | Positions Amazon Q Developer as a generative AI–powered assistant for software development across the software development lifecycle. | Not verified in this capture |
| Kiro grades A/A/A/A/A/A | Professional developers on software teams, businesses and enterprises; plans sold per user for organisations adopting Kiro across a team. | Free Tier: $0 per month (includes 50 credits per pricing FAQ) · Kiro Pro: $20 per user / month · Kiro Pro+: $40 per user / month · Kiro Pro Max: $100 per user / month · Kiro Power: $200 per user / month · Kiro for enterprise: Contact sales / enterprise packaging | Yes | Positions Kiro as moving beyond AI coding to agentic engineering via spec-driven development, correctness checks and parallel agents across IDE, CLI, web and related surfaces. | Claude Sonnet 5.5 is now available in Kiro |
| Sourcegraph (enterprise code search and Deep Search) grades A/A/A/A/A/D | Leading engineering teams / enterprise scale (pricing page: built for enterprise scale; give engineers and agents deep codebase context). | Enterprise plan: Starting at $16K minimum annual contract. Includes credits for AI features; scales with team size | No | One platform to understand, oversee, and evolve the world's most complex codebases, including search and navigation and AI-powered Deep Search. | Not verified in this capture |
| JetBrains AI grades A/A/A/A/A/D | Developers and teams using JetBrains IDEs, browser and CLI surfaces to build software with agents (including Claude Agent, Codex, Junie, Copilot, OpenCode and ACP-connected agents). | AI Trial (tab: Individual use): Free (10 AI Credits per trial period) (USD) · AI Free (tab: Individual use): Free (3 AI Credits per 30-days) (USD) · AI Pro (tab: Individual use): $10 USD · AI Ultimate (tab: Individual use): $30 USD · AI Trial (tab: Organizations (per-user)): Free (20 AI Credits per trial period) (USD) · AI Free (tab: Organizations (per-user)): Free (3 AI Credits per 30-days) (USD) · AI Pro (tab: Organizations (per-user)): $20 USD · AI Ultimate (tab: Organizations (per-user)): $60 USD · AI Enterprise (tab: Organizations (per-user)): $60 USD | Yes | Positions JetBrains Air as one system for building software with agents, working with agents developers already use and connecting via ACP. | Not verified in this capture |
Dates are announcement or publication dates as shown on the cited page.
Market gaps
Grade C only — labelled interpretation, not established fact.
- GRADE C · INTERPRETATION
Interpretation: Multiple official pages describe usage allowances, credits or quotas that reset on different cadences and consume differently by model or task. From public pricing copy alone, comparing “how far included usage goes” across assistants is difficult without running a controlled evaluation.
Recommendations
Illustrative only — not advice tailored to a specific buyer.
- Illustrative: Do not treat undercutting the observed $10–$20 entry / $100–$200 power ladder as a durable wedge on the basis of these public price lists alone. (from F1)
- Illustrative: Do not treat new-model access as a lasting differentiator: two vendors announced a Claude 5.5 model within a week of each other. (from F2)
- Illustrative: When an incumbent retires a standalone product for a newer agentic one, compare against the successor product (here Kiro), not the retired one. (from F3)
Evidence key
Grade A means stated directly on an official vendor page. It is not independent corroboration; that is Grade B, and this sample claims none.
- Grade A
- Direct from an official page53 claims in this file
- Grade B
- Corroborated by two or more independent sources0 claims in this file
- Grade C
- Inference / interpretation4 claims in this file
- Grade D
- Unknown / not verified in this capture4 claims in this file
Source appendix
Every URL used in the data file, with retrieval date. Count: 21.
- https://aws.amazon.com/q/developer/
Retrieved 2026-10-08 · grades A, D
- https://aws.amazon.com/q/developer/pricing/
Retrieved 2026-10-08 · grades A
- https://claude.com/pricing
Retrieved 2026-10-08 · grades A
- https://cursor.com
Retrieved 2026-10-08 · grades A
- https://cursor.com/changelog/remote-control-local-agents
Retrieved 2026-10-08 · grades A
- https://cursor.com/docs/models-and-pricing
Retrieved 2026-10-08 · grades A
- https://cursor.com/pricing
Retrieved 2026-10-08 · grades A
- https://devin.ai
Retrieved 2026-10-08 · grades A
- https://devin.ai/blog/claude-haiku-5-5
Retrieved 2026-10-08 · grades A
- https://devin.ai/pricing
Retrieved 2026-10-08 · grades A
- https://docs.github.com/en/copilot/get-started/plans
Retrieved 2026-10-08 · grades A
- https://github.blog/news-insights/company-news/github-copilot-individual-plans-introducing-flex-allotments-in-pro-and-pro-and-a-new-max-plan/
Retrieved 2026-10-08 · grades A
- https://github.com/features/copilot/plans
Retrieved 2026-10-08 · grades A
- https://kiro.dev
Retrieved 2026-10-08 · grades A
- https://kiro.dev/blog/sonnet-5-5/
Retrieved 2026-10-08 · grades A
- https://kiro.dev/pricing
Retrieved 2026-10-08 · grades A
- https://sourcegraph.com/pricing
Retrieved 2026-10-08 · grades A, D
- https://www.anthropic.com/claude-code
Retrieved 2026-10-08 · grades A
- https://www.anthropic.com/news
Retrieved 2026-10-08 · grades D
- https://www.jetbrains.com/ai/
Retrieved 2026-10-08 · grades A, D
- https://www.jetbrains.com/help/ai-assistant/licensing-and-subscriptions.html
Retrieved 2026-10-08 · grades A
Brief us
Need a decision-grade landscape for your market? Send a brief — we review every request before discussing an engagement.
Brief us