GEM-4C 558 -4%
GPT-61S 555 -3.8%
CL-SN55 553 -3.8%
CL-OP55 551 -4%
GPT-61TERR 547 -3.5%
GEM-4CM 546 -4%
CL-OP55EH 542 -3.9%
GPT-61SOLXH 541 -3.6%
GPT-61SOLH 527 -3.7%
CL-SN55XH 523 -3.5%
GEM-4CL 520 -3.9%
GPT-61TERRXH 507 -3.4%
CL-OP55H 506 -4%
GPT-61SOLM 500 -3.5%
KIMI-K31P 499 -3.5%
GEM-4AR 498 -3.9%
GPT-61TERRH 490 -3.2%
CL-OP55M 483 -3.6%
MUSE-WONDFUL 479 -3.8%
CL-SN55H 467 -3.3%
CL-HK55 463 -2.9%
GPT-61TERRM 462 -3.1%
GPT-6A 451 -4.7%
MUSE-WOND 451 -3.6%
CL-HK55XH 450 -3%
GPT-6ASTRXH 440 -4.8%
GROK-47 440 -3.1%
GROK-47H 438 -2.9%
GPT-6ASTRH 437 -4.8%
DSK-V-340615 436 -2.9%
QWEN-38X 428 -2.9%
GPT-61SOLL 423 -3.2%
GLM-53 420 -3%
MIMO-V26P 420 -2.6%
CL-HK55H 419 -2.8%
LING-31F1019 417 -2.6%
GEM-38F 416 -2.8%
GPT-6ASTRM 415 -4.6%
MMAX-M31F 414 -2.8%
GEM-38FM 412 -2.8%
GPT-61TERRNO 411 -2.8%
GROK-47L 410 -2.8%
DSK-V-3E48BD 410 -2.6%
GPT-6L 407 -2.6%
GLM-53F 407 -2.6%
STEP-5 405 -2.6%
CL-FABL51XH 403 -4.7%
GPT-61TERRL 403 -2.9%
MIMO-V26F 401 -2.4%
CL-FABL51 399 -4.5%
CL-SN55M 395 -2.9%
CL-OP55L 393 -3.4%
MIST-LARG4 390 -2.7%
CL-FABL51H 386 -4.5%
CL-HK55M 386 -2.3%
GPT-6LUNAXH 377 -2.6%
MUSE-SPAROPE 375 -2.6%
MIST-LARG4M 371 -2.4%
DSK-V4P 365 -2.4%
GPT-6LUNAH 365 -2.1%
GLM-53L 364 -2.4%
CL-SN55L 362 -2.7%
GEM-38FL 361 -2.4%
GPT-6ASTRL 359 -4.5%
CL-FABL51M 359 -4.5%
QWEN-827B 349 -2.2%
CL-HK55L 347 -2%
MMAX-M3 345 -2%
GPT-6LUNAM 344 -2.3%
DSK-V-02811D 344 -2%
K2-HORI375BA 342 -2%
REFL-BEAM 338 -2%
QWEN-827BM 332 -1.8%
CL-FABL51L 329 -4.4%
QWEN-827BL 320 -1.8%
NEMO-35SUPEV 313 -1.9%
INKL 295 -1.7%
NVID-NEMO3UL 295 -1.3%
GEM-35FLITE 295 -1.3%
GPT-6LUNAL 287 -1.4%
GPT-6LUNANON 286 -1%
NANB-333B 285 -2.7%
QWEN-827BNON 282 -1.4%
COMM-APLUS 266 -1.1%
MUSE-GLIM 263 -1.1%
NEMO-35LIGH 261 -0.8%
GRAN-428B 258 -0.8%
MIST-SMAL4 252 -0.4%
INKL-SM 235 -2.5%
GEMM-431B 225 —
GEMM-4E4B 225 —
GEMM-4E4BNON 225 —
LING-30TINY 225 —
MINI-33B 225 —
NORT-MCODE 225 —
NOVA-MICR 225 —
NVID--ADE3B4 225 —
QWEN-54B 225 —
QWEN-54BNONR 225 —
TINY-AYAGLOB 225 —
GEM-4C 558 -4%
GPT-61S 555 -3.8%
CL-SN55 553 -3.8%
CL-OP55 551 -4%
GPT-61TERR 547 -3.5%
GEM-4CM 546 -4%
CL-OP55EH 542 -3.9%
GPT-61SOLXH 541 -3.6%
GPT-61SOLH 527 -3.7%
CL-SN55XH 523 -3.5%
GEM-4CL 520 -3.9%
GPT-61TERRXH 507 -3.4%
CL-OP55H 506 -4%
GPT-61SOLM 500 -3.5%
KIMI-K31P 499 -3.5%
GEM-4AR 498 -3.9%
GPT-61TERRH 490 -3.2%
CL-OP55M 483 -3.6%
MUSE-WONDFUL 479 -3.8%
CL-SN55H 467 -3.3%
CL-HK55 463 -2.9%
GPT-61TERRM 462 -3.1%
GPT-6A 451 -4.7%
MUSE-WOND 451 -3.6%
CL-HK55XH 450 -3%
GPT-6ASTRXH 440 -4.8%
GROK-47 440 -3.1%
GROK-47H 438 -2.9%
GPT-6ASTRH 437 -4.8%
DSK-V-340615 436 -2.9%
QWEN-38X 428 -2.9%
GPT-61SOLL 423 -3.2%
GLM-53 420 -3%
MIMO-V26P 420 -2.6%
CL-HK55H 419 -2.8%
LING-31F1019 417 -2.6%
GEM-38F 416 -2.8%
GPT-6ASTRM 415 -4.6%
MMAX-M31F 414 -2.8%
GEM-38FM 412 -2.8%
GPT-61TERRNO 411 -2.8%
GROK-47L 410 -2.8%
DSK-V-3E48BD 410 -2.6%
GPT-6L 407 -2.6%
GLM-53F 407 -2.6%
STEP-5 405 -2.6%
CL-FABL51XH 403 -4.7%
GPT-61TERRL 403 -2.9%
MIMO-V26F 401 -2.4%
CL-FABL51 399 -4.5%
CL-SN55M 395 -2.9%
CL-OP55L 393 -3.4%
MIST-LARG4 390 -2.7%
CL-FABL51H 386 -4.5%
CL-HK55M 386 -2.3%
GPT-6LUNAXH 377 -2.6%
MUSE-SPAROPE 375 -2.6%
MIST-LARG4M 371 -2.4%
DSK-V4P 365 -2.4%
GPT-6LUNAH 365 -2.1%
GLM-53L 364 -2.4%
CL-SN55L 362 -2.7%
GEM-38FL 361 -2.4%
GPT-6ASTRL 359 -4.5%
CL-FABL51M 359 -4.5%
QWEN-827B 349 -2.2%
CL-HK55L 347 -2%
MMAX-M3 345 -2%
GPT-6LUNAM 344 -2.3%
DSK-V-02811D 344 -2%
K2-HORI375BA 342 -2%
REFL-BEAM 338 -2%
QWEN-827BM 332 -1.8%
CL-FABL51L 329 -4.4%
QWEN-827BL 320 -1.8%
NEMO-35SUPEV 313 -1.9%
INKL 295 -1.7%
NVID-NEMO3UL 295 -1.3%
GEM-35FLITE 295 -1.3%
GPT-6LUNAL 287 -1.4%
GPT-6LUNANON 286 -1%
NANB-333B 285 -2.7%
QWEN-827BNON 282 -1.4%
COMM-APLUS 266 -1.1%
MUSE-GLIM 263 -1.1%
NEMO-35LIGH 261 -0.8%
GRAN-428B 258 -0.8%
MIST-SMAL4 252 -0.4%
INKL-SM 235 -2.5%
GEMM-431B 225 —
GEMM-4E4B 225 —
GEMM-4E4BNON 225 —
LING-30TINY 225 —
MINI-33B 225 —
NORT-MCODE 225 —
NOVA-MICR 225 —
NVID--ADE3B4 225 —
QWEN-54B 225 —
QWEN-54BNONR 225 —
TINY-AYAGLOB 225 —

Delayed ticker data. Last calculated 5 Nov 2026, 11:04 pm Sydney time.

← Back to feed

Munder Difflin Is the Free Local Multi-Agent Harness Developers Are Actually Using

A tool called Munder Difflin—named after The Office’s fictional paper company—hit Hacker News front page this week, surfacing as the strongest community signal yet for developer demand in self-hosted multi-agent orchestration.

The project, available at github.com/chaitanyagiri/munder-difflin, is a local-first agent harness that coordinates multiple AI instances in parallel. It runs against API subscriptions the user already holds—Claude, GPT-5.x, or others—working within their existing rate limits rather than requiring a new managed service.

What It Does

Munder Difflin organizes agents as an “office of clones”: identical or specialized agent instances given discrete tasks, connected through shared memory and a task board. An org-chart-style hierarchy lets one coordinator agent delegate to worker agents, enabling persistent autonomous runs without a human in the loop.

Features at current release:

  • Parallel agent coordination across any API with existing credentials
  • Shared memory layer so agents can read each other’s output and context
  • Task board for distributing and tracking work items
  • Org-chart delegation: coordinator → specialist hierarchy
  • Long-running autonomous mode: tasks can run unattended for extended periods
  • MIT license, Windows/macOS/Linux support

The Cost Argument

Commercial multi-agent platforms have a predictable pricing structure: model API costs plus a platform fee. OpenAI Codex charges per hour of compute. Anthropic Managed Agents prices at $0.08 per agent-hour. LangGraph Cloud adds orchestration overhead on top of model costs.

Munder Difflin removes the platform layer entirely. If a developer has an existing API subscription, running an office of agents costs only the token charges from those models—nothing additional.

That positioning is what the HN front-page placement reflects. The multi-agent story has been dominated by enterprise-facing managed services. Munder Difflin’s free, local approach targets developers who already have API access and don’t want another SaaS bill.

Context

Multi-agent orchestration has moved fast in 2026. OpenAI’s Codex V2 added delegation from Sol to Luna subagents in August. Anthropic Managed Agents reached public beta. Both are managed, metered services. Open-source frameworks like LangGraph and AutoGen have existed for years but carry engineering overhead.

Munder Difflin’s single-install, no-new-subscription model sits between the two extremes—easier to start than raw frameworks, cheaper than managed platforms.

HN discussion: news.ycombinator.com/item?id=49398152