Thinking Machines Releases Inkling 2 Small, Open Weights Next Week

Thinking Machines, the company founded by Mira Murati, released its second-generation model Inkling 2 Small on November 6 and announced progress on its "model factory." Following Inkling Small, the release focuses on long-horizon autonomous tasks and post-trainability.

Confirmed

• Architecture and capabilities: a MoE architecture with 594B total parameters and 17B activated, supporting image input and variable effort; built for long-horizon autonomous tasks, able to visually inspect its work and iterate, and adaptable to different agent frameworks and environments
• Benchmarks: scored 31 on the Artificial Analysis Intelligence Index v5.0.1, up 18 points from the previous-generation Inkling Small
• Availability: fine-tuning and inference are now open on the Tinker platform—users can fine-tune it themselves or let the model autonomously orchestrate the entire fine-tuning workflow; full weights will be open-sourced soon
• Training efficiency: per engineering lead cHHillee, cited in a repost, pretraining took less than 3 days; developer simonguozirui reported that 75 steps of fine-tuning on Tinker matched the previous generation's final score, with fine-tuning efficiency far ahead across multiple post-training scenarios using the same data and algorithms
• Positioning: if the weights are open-sourced as planned, it would become the strongest open-weight model outside China

Why it matters

• "Post-trainability" is the core selling point: the release emphasizes developer efficiency in fine-tuning over conventional benchmark scores, which directly affects the practical value of open-weight models
• Reaching this performance with less than 3 days of pretraining (if accurate) showcases the rapid iteration of Thinking Machines' "model factory," suggesting notably faster release cadence going forward
• If it truly becomes the strongest open-weight model outside China after release, it would reshape the current open-source model landscape with direct impact on developers who rely on open weights

Not yet confirmed

• Pricing, detailed training specifics, and a full technical report have not been disclosed
• The "strongest open-weight model" claim rests on a single Artificial Analysis index and requires more third-party evaluation once the weights are actually released

2026-11-06 ~ 2026-11-06 · 34 related posts

Full story(4 episodes)→

• Episode 1: Thinking Machines Releases Inkling 2 Small, Open Weights Next Week(2026-11-06, 34 posts)
• Episode 2: Inkling 2 Small tops Design Arena open-weights ranking(2026-11-06, 3 posts)
• Episode 3: Inkling 2 Small generates full stock diligence memo for $0.69(2026-11-06, 2 posts)
• Episode 4: Thinking Machines Releases Inkling 2 Small as Strongest Open-Weight Model Outside China(2026-11-06, 3 posts)

• [source] Thinking Machines' Inkling 2 Small targets long-horizon tasks with visual self-checking — thinkymachines · 2026-11-06
• Thinking Machines launches Inkling 2 Small: 594B MoE, AA Intelligence Index score of 31 — stephenroller · 2026-11-06
• Mira Murati's Thinking Machines Releases Inkling 2 Small, Scoring 31 on AA Index — saurabh_garg67 · 2026-11-06
• [source] Thinking Machines' Inkling 2 Small scores 31 on AA Intelligence Index, planned weight release would top non-China open models — ArtificialAnlys · 2026-11-06
• [source] Inkling 2 Small launches on Tinker, benchmarked for post-trainability — simonguozirui · 2026-11-06
• Thinking Machines releases Inkling 2 Small, pretrained in under 3 days — SonglinYang4 · 2026-11-06
• Thinking Machines ships Inkling 2 Small as its model factory advances — VictoriaLinML · 2026-11-06
• Thinking Machines' Inkling 2 Small Hits #4 on Design Arena Frontend Leaderboard — MuCai7 · 2026-11-06
• Inkling 2 Small released: Thinking Machines' new 594B-A17B model — spiwar · 2026-11-06
• Thinking Machines announces Inkling 2 Small (594B-A17B), open weights next week — tkrandomness · 2026-11-06
• Inkling 2 Small team member: "approaching frontier", soliciting user feedback — LiTianleli · 2026-11-06
• Thinking Machines Ships Inkling 2 Small, Takes #3 on SWE-bench Multimodal — ziqiao_ma · 2026-11-06
• Inkling 2 Small ships with focus on midtraining; team hiring — LiTianleli · 2026-11-06
• Inkling 2 Small tops cost-performance frontier at $0.34/task on sudo L7 — echen · 2026-11-06
• Thinky Machines launches Inkling 2 Small, scoring 31 on AA Intelligence Index — LiTianleli · 2026-11-06
• Inkling 2 Small Launches: 594B Params Native NVFP4, Open Weights Next Week — LiTianleli · 2026-11-06
• Tinker launches Inkling 2 Small, built around customizability with launch-blocking eval — simonguozirui · 2026-11-06
• Inkling 2 Small matches models 8-13x pricier at just $0.34 per task — echen · 2026-11-06
• Pretrained in Just 68 Hours: Inkling 2 Small Scores 31 on AA Intelligence Index — simonguozirui · 2026-11-06
• Thinking Machines releases Inkling 2 Small, scoring 31 on AA Intelligence Index — simonguozirui · 2026-11-06
• Inkling-Small was harder to post-train than Qwen; Inkling 2 targets that — LiTianleli · 2026-11-06
• Thinking Machines launches Inkling 2 Small: 594B MoE scoring 31 on AA Index, Apache 2.0 weights next week — MohitIyyer · 2026-11-06
• Inkling 2 Small beats Kimi K3.1 on code loss with just 12T pretrain tokens — teortaxesTex · 2026-11-06

11 near-duplicate retellings: simonguozirui · LiTianleli · LiTianleli · liliyu_lili · echen · LiTianleli · simonguozirui · jonlachman · LiTianleli · VictoriaLinML · SonglinYang4