GPT-56T 861 —
MUSE-SPK 837 +0.2%
GPT-56SC 790 -4.6%
GLM-5 781 -0.4%
CL-OP55X 780 -5.1%
GROK-46H 780 -5.1%
QWEN-38X 748 -9.2%
GPT-6A 743 -9.4%
KIMI-K3X 742 —
CL-FAB5H 698 -6.1%
CL-OP5H 675 -6.2%
GEM-38FH 672 -0.7%
CL-OP5X 670 -5.5%
CL-OP55H 668 —
CL-OP46H 657 -5.9%
CL-OP47H 648 -6.1%
GPT-56S 618 -0.6%
GEM-37FH 610 -7.2%
GEM-36FH 593 —
CL-OP48H 588 —
CL-OP47 581 -0.2%
GEM-35FH 580 —
GPT-55H 541 -7%
INKL 531 —
GEM-31P 512 -0.2%
CL-OP46 498 +0.4%
GEM-3P 498 -0.2%
CL-OP48 492 +0.4%
GPT-52 464 —
GPT-55 423 —
GPT-56T 861 —
MUSE-SPK 837 +0.2%
GPT-56SC 790 -4.6%
GLM-5 781 -0.4%
CL-OP55X 780 -5.1%
GROK-46H 780 -5.1%
QWEN-38X 748 -9.2%
GPT-6A 743 -9.4%
KIMI-K3X 742 —
CL-FAB5H 698 -6.1%
CL-OP5H 675 -6.2%
GEM-38FH 672 -0.7%
CL-OP5X 670 -5.5%
CL-OP55H 668 —
CL-OP46H 657 -5.9%
CL-OP47H 648 -6.1%
GPT-56S 618 -0.6%
GEM-37FH 610 -7.2%
GEM-36FH 593 —
CL-OP48H 588 —
CL-OP47 581 -0.2%
GEM-35FH 580 —
GPT-55H 541 -7%
INKL 531 —
GEM-31P 512 -0.2%
CL-OP46 498 +0.4%
GEM-3P 498 -0.2%
CL-OP48 492 +0.4%
GPT-52 464 —
GPT-55 423 —
← Back to feed

IBM Bob Launches Globally: Enterprise AI Coding Agent Claims 45% Productivity Gains Across 80,000 Employees

IBM shipped IBM Bob to general availability on April 28, completing a rollout that started with 100 internal developers in June 2025 and has since reached 80,000 IBM employees worldwide. The tool positions itself against Cursor, GitHub Copilot, and Claude Code — but with a compliance and auditability frame that none of the others lead with.

What it does

Bob’s differentiation from point-solution coding assistants is scope. It covers the full software development lifecycle from discovery and planning through design, coding, testing, deployment, and legacy modernization. It is not just a code completion layer.

Key capabilities at launch:

  • Multi-model orchestration: Routes tasks across Anthropic Claude, Mistral open-source models, IBM Granite SLMs, and fine-tuned specialist models for code reasoning, security scanning, and next-edit prediction. Simpler tasks go to lighter models; complex tasks escalate automatically.
  • BobShell CLI: Every agentic action is logged in real time, producing a full audit trail from start to finish. Designed for teams that need to trace AI-generated code changes back to their origin.
  • Governance baked in: Prompt normalization, PII/sensitive data scanning, real-time policy enforcement, and AI red-teaming built into the workflow — not added as a compliance layer afterward.
  • Intelligent modernization: IBM cites Blue Pearl, a cloud consulting firm, which used Bob to compress a typical 30-day Java upgrade into 3 days, saving 160+ engineering hours.
  • Pass-through pricing: IBM is not marking up model costs. Organizations see what the underlying models cost per task.

Internal evidence

The 80,000-employee internal deployment is IBM’s strongest claim. Surveyed users self-report an average 45% productivity gain across modernization, security, and new development. The IBM Instana team reported 70% reduction in time on selected tasks, equaling 10 hours per week. The IBM Maximo team compressed multi-day code refactoring to hours.

Self-reported figures carry the usual caveats, but the deployment scale gives IBM a dataset that no startup entering the enterprise market can match at launch.

Market context

IBM Bob enters a crowded field. Cursor hit $2B ARR. GitHub Copilot is moving to token-based pricing. Anthropic launched Claude Managed Agents in the cloud. AWS AgentCore ships persistent filesystem support.

IBM’s counter-positioning is the enterprise stack: governance, auditability, on-premises roadmap for data residency, and a model routing layer that does not assume teams want to be locked to a single frontier provider. That framing targets the regulated industries — financial services, healthcare, government — where Claude Code Pro and Cursor Pro have not yet won.

Pricing and availability

SaaS, available now at bob.ibm.com. Individual and enterprise plans; a 30-day free trial is included. On-premises deployment is on the roadmap for organizations with data residency or regulatory requirements. IBM has not published per-seat or token pricing publicly.