---CLAUDE---
score: 88
trend: down
change: -1
PRO: Opus 4.7 retakes coding lead — SWE-bench Verified 87.6%, Pro 64.3%, beats GPT-5.4 and Gemini 3.1 Pro
PRO: Ramp data shows Anthropic at 30.6% of AI-paying business customers vs OpenAI 35.2%, crossover within two months
PRO: White House opens civilian Mythos lane — Wiles and Bessent met Amodei April 17 despite Pentagon blacklist
CON: Three outages between April 10-15, 90-day API uptime at 98.95% vs 99.99% enterprise standard
CON: Enterprise billing unbundled to per-token — heavy Claude Code users face material bill increases at renewal
CON: Retool CEO David Hsu publicly moved coding workloads to OpenAI over reliability
---GEMINI---
score: 81
trend: up
change: +2
PRO: Crosses ChatGPT in the meter — $2/$12 per million tokens runs 2.14x cheaper than Opus 4.7 at comparable SWE-bench
PRO: Gemini handles roughly 45% of Fortune 500 enterprise AI queries inside Workspace deployments
PRO: Cloud Next '26 opens April 22 with agentic, governance, security roadmap reveals
PRO: Mariner Computer Use shipped, Dropbox federated data store GA in Gemini Enterprise
CON: Lost coding benchmark crown to Opus 4.7 this cycle — marketing dent, not economic
---CHATGPT---
score: 79
trend: down
change: -2
PRO: GPT-5.4 still leads terminal work and computer use, GPT-5.3 Instant Mini upgrades fallback tier
PRO: Enterprise now 40%+ of revenue, on track for parity with consumer by year-end
PRO: Hiro Finance acquihire adds personal CFO layer, London UK hub locks in 544 seats
CON: Three senior exits in one day — Weil, Peebles, Narayanan — as Sora shuts April 26 and science team folds into Codex
CON: Enterprise AI spend share fell from 50% to 27% as Anthropic climbed to 40%
CON: CRO memo war against Anthropic and $852B valuation under investor scrutiny read as defensive
---MISTRAL---
score: 69
trend: up
change: +2
PRO: Connectors launched April 15 — MCP, direct tool calling, human-in-the-loop approval flows exposed via API
PRO: Le Chat now integrates 20+ enterprise platforms including Databricks, Snowflake, GitHub, Atlassian, Stripe
PRO: Accenture strategic collaboration and French DINUM sovereignty mandate continue to channel deal flow
CON: North American Fortune 500 penetration remains thin against the hyperscaler-native options
---GROK---
score: 35
trend: down
change: -5
PRO: Pursuing FedRAMP High authorization sponsored by USDA — multi-year process, not near-term
PRO: Grok Computer private beta opens agentic workload positioning, Colossus 2 scaling continues
CON: Apple threatened App Store removal over non-consensual sexualized deepfakes, NBC reports problem persists
CON: xAI sued Colorado April 9 to block AI anti-discrimination law — hostile posture toward compliance frameworks
CON: SpaceX IPO underwriters required to buy Grok subscriptions as condition of participation
---DEEPSEEK---
score: 13
trend: down
change: -1
PRO: $300M raise at $10B valuation — first-ever outside funding, rationalizes V4's $500M+ training cost
PRO: V4 late-April launch signals with 1T parameters, 1M context, Apache 2.0 under Huawei Ascend optimization
CON: V4 missed another deadline — API still serves V3.2, multiple prior slips
CON: Huawei Ascend path decouples further from Western accelerator stacks, blocks regulated deployment
CON: Pentagon pressure to designate as "Chinese military firm" keeps US VC and enterprise wary
IMPLICATOR