--- title: >- Open weights finally became deployable artifacts — and the closed stack answered with post-training and rate cards, not new flagships publication: The Model Pulse slug: 2026-W31 issueNumber: 15 isoYear: 2026 isoWeek: 31 cadence: weekly publishedAt: '2026-08-01' periodLabel: Week 31 of 2026 canonicalUrl: https://brianletort.ai/industry/models/2026-W31 pdfUrl: https://brianletort.ai/downloads/model-pulse-2026-W31.pdf schemaVersion: 2026.05.02 treeDelta: added: - inkling-small updated: - kimi-k3 - deepseek-v4-flash frontierMovements: - modelId: null name: GPT-5.6 Luna / Terra price cut and Sol Fast mode vendor: OpenAI releaseDate: '2026-07-30' tier: frontier architecture: reasoning - modelId: null name: DeepSeek V4-Flash-0731 vendor: DeepSeek releaseDate: '2026-07-31' tier: frontier architecture: moe - modelId: null name: GPT-5.6 Sol harness memory on ARC-AGI-3 vendor: OpenAI releaseDate: '2026-07-29' tier: reasoning architecture: agentic - modelId: null name: Qwen3.7 Flash vendor: Alibaba / Qwen releaseDate: '2026-07-27' tier: specialist architecture: multimodal openWeights: - modelId: kimi-k3 name: Kimi K3 vendor: Moonshot AI releaseDate: '2026-07-27' tier: open_frontier architecture: moe - modelId: null name: Inkling-Small vendor: Thinking Machines Lab releaseDate: '2026-07-30' tier: open_frontier architecture: moe - modelId: null name: MiniMax H3 vendor: MiniMax releaseDate: '2026-07-31' tier: specialist architecture: multimodal architecturePatterns: - Post-training and harness policy move scores without new weights - Revenue-tiered open licenses turn legal review into the deployment gate - Serving stacks ship in the same week as the open MoE class - Near-iso Intelligence at a fraction of active parameters benchmarks: - Artificial Analysis Intelligence Index — DeepSeek V4 Flash - Artificial Analysis Intelligence Index — Inkling-Small - ARC-AGI-3 public set (vendor harness, not official board) - DeepSeek Terminal-Bench 2.1 (vendor harness vs independent) scorecardAsOf: '2026-08-01' vendorSignals: - vendor: OpenAI date: '2026-07-30' signal: >- GPT-5.6 Luna cut 80% to $0.20/$1.20 and Terra 20% to $2/$12; Sol held at $5/$30; Fast mode replaces Priority Processing - vendor: Moonshot AI date: '2026-07-27' signal: >- Kimi K3 License ships with weights: MIT-like until MaaS revenue exceeds $20M trailing twelve months - vendor: Google DeepMind date: '2026-07-31' signal: >- Gemini 3.5 Pro still not generally available as the standing Jul 31 prediction deadline expires - vendor: MiniMax date: '2026-07-31' signal: >- H3 launches with an explicit promise to open weights in the coming days; no public HF or GitHub artifacts as of Aug 1 - vendor: vLLM project date: '2026-07-27' signal: >- vLLM v0.26.0 ships Inkling family support, NVFP4, MTP speculative decoding, and DeepSeek-V4 serving optimizations --- # Open weights finally became deployable artifacts — and the closed stack answered with post-training and rate cards, not new flagships *The Model Pulse · Issue 15 · Week 31 of 2026 · Published 2026-08-01* ## Big Read Moonshot's Jul 27 Kimi K3 weight drop converts last week's promissory open-frontier watch item into a procurement fact: a 2.8T MoE with 104B active parameters, 896 experts selecting 16, hybrid 69 KDA + 24 Gated MLA layers, 1,048,576 context, and MoonViT-V2 — under a revenue-tiered Kimi K3 License that triggers a separate commercial deal for MaaS above $20M trailing-twelve-month revenue. Self-host evaluation can now start on artifacts; license diligence is the gate, not model quality. No closed frontier model shipped. The closed stack competed on price and post-training instead. OpenAI cut GPT-5.6 Luna 80% to $0.20/$1.20 and Terra 20% to $2/$12 on Jul 30 while holding Sol at $5/$30 and replacing Priority Processing with Fast mode at 2x token price for up to ~2.5x speed. DeepSeek put V4-Flash-0731 into public beta on Jul 31 on the same 284B/13B architecture and unchanged $0.14/$0.28 pricing; Artificial Analysis independently scored it +10 Intelligence Index points to 50. Thinking Machines released Inkling-Small the same day as the Luna cut — 276B/12B active, Apache 2.0, independent AA Index 40 versus Inkling's 41 — with a context-window conflict to resolve (vendor card up to 1M; AA article lists 256K). The week's largest vendor-claimed capability jump did not need new weights either: OpenAI reported ARC-AGI-3 public-set scores moving from 13.3% to 38.3% on GPT-5.6 Sol by retaining reasoning between tool calls and compacting context instead of truncating, with roughly 6x fewer output tokens. That is a harness result, not an official leaderboard reprint — and no qualifying LMArena, SWE-bench, or ARC official board moves landed in-window. Gemini 3.5 Pro missed its Jul 31 GA prediction. MiniMax H3 promised open weights and had not shipped them as of Aug 1. Net/net: route on measured Index and task cost, treat open as license-plus-weights, and re-test agents against memory policy before buying a bigger model. ## Tree delta Inkling-Small joins the tree as the strongest US open-weights efficiency result to date, reaching an independent Intelligence Index of 40 against Inkling's 41 with under a third of the parameters. Kimi K3 is reclassified from closed to open weights on the July 27 artifacts and license text, with its active-parameter count corrected from a community estimate to a confirmed 104B. DeepSeek V4 Flash updates for the 0731 re-post-train checkpoint that gained 10 independent Index points on unchanged architecture. **Added (1):** `inkling-small`. **Updated (2):** `kimi-k3`, `deepseek-v4-flash`. *Kimi K3 is an update rather than an addition: the row has existed since W29, and this week's material change is openness, license terms, and a corrected active-parameter count of 104B against the earlier community-estimated ~50B. Qwen3.7 Flash is held out of the tree pending a first-party model card, and MiniMax H3 is deferred until its promised weights actually ship. Both are covered in the cards.* ## Frontier movements ### GPT-5.6 Luna / Terra price cut and Sol Fast mode **OpenAI** · 2026-07-30 · frontier · reasoning _Luna falls 80% to $0.20/$1.20, Terra 20% to $2/$12, Sol holds $5/$30, and Fast mode sells ~2.5x speed at 2x token price_ Re-baseline high-volume routing immediately: Luna undercuts most Western cheap tiers on sticker price while Sol remains the premium band against Opus 5. Treat Fast mode as a purchasable latency tier rather than a separate model, and instrument per-tier completion rates before moving fleet volume on list price alone. The AI Stack Weekly owns the fixed-workload spread calculation; this Pulse stays on the architecture and routing implication. Source: [OpenAI; Unite.AI; VentureBeat; The Decoder](https://openai.com/index/advancing-the-price-performance-frontier-with-gpt-5-6/). ### DeepSeek V4-Flash-0731 **DeepSeek** · 2026-07-31 · frontier · moe _Same 284B/13B MoE and $0.14/$0.28 pricing, re-post-trained into public beta; independent AA Index jumps 10 points to 50_ Drop this into existing Flash routes as a cheap-tier agent upgrade without a size or price change — but do not treat it as V4-Pro GA, and pin the harness when comparing Terminal-Bench (vendor-claimed 82.7 max effort versus Artificial Analysis 79%). Architecture and model ID stay deepseek-v4-flash; Pro, app, and web remain unchanged pending the promised official V4-Pro. Source: [DeepSeek API changelog; Artificial Analysis (independent)](https://api-docs.deepseek.com/updates/). ### GPT-5.6 Sol harness memory on ARC-AGI-3 **OpenAI** · 2026-07-29 · reasoning · agentic _Vendor-reported public-set score triples from 13.3% to 38.3% via retained reasoning plus context compaction on the same model_ Treat the ARC-AGI-3 jump as vendor-claimed on a single harness, not an official leaderboard reprint: OpenAI reports roughly 6x fewer output tokens when compaction replaces rolling truncation on the Responses API, and no qualifying LMArena or SWE-bench official moves landed in-window. Agent Techniques Weekly owns the method and the reproduction caveat. Source: [OpenAI (vendor-reported harness result)](https://openai.com/index/how-two-settings-tripled-our-arc-agi-3-scores/). ### Qwen3.7 Flash **Alibaba / Qwen** · 2026-07-27 · specialist · multimodal _Quiet low-cost multimodal API at reported $0.03/$0.13 under 32K and 1M context — no lab blog, no weights, no public benchmark table_ Useful as a fleet-cost floor candidate for multimodal agents, but treat specs and pricing as secondary until a first-party model card is pinned. Sources are OpenRouter listing dates and technical digests citing a QwenCloud changelog; no qwenlm.github.io launch post was found in-window. Source: [eesel.ai; BenchLM (secondary; no primary lab announcement)](https://benchlm.ai/models/qwen3-7-flash). ## Open weights ### Kimi K3 `kimi-k3` **Moonshot AI** · 2026-07-27 · open_frontier · moe _Full weights ship: 2.8T MoE, 104B active, 896/16 experts, hybrid 69 KDA + 24 Gated MLA, 1,048,576 context, revenue-tiered license_ Start the self-host evaluation now — GitHub and Hugging Face artifacts landed Jul 27 — but send the Kimi K3 License to counsel before any commercial hosting plan: MaaS above $20M aggregate trailing-12-month revenue needs a separate deal, with attribution triggers above 100M MAU or $20M monthly revenue. Independent AA Index reference remains 57; vendor-claimed table includes GPQA Diamond 93.5 and DeepSWE 67.5. Hosted API stays $3/$15 (cached input $0.30). Source: [MoonshotAI/Kimi-K3 GitHub; Hugging Face; license text; Unite.AI](https://huggingface.co/moonshotai/Kimi-K3). ### Inkling-Small **Thinking Machines Lab** · 2026-07-30 · open_frontier · moe _276B/12B-active multimodal MoE under Apache 2.0; independent AA Index 40 vs Inkling 41 at under a third the parameters_ Strongest US open-weights efficiency play this week for teams that cannot absorb K3-class cluster scale: BF16 needs ~600 GB aggregate VRAM, NVFP4 ~180 GB, with vLLM v0.26.0 shipping same-week Inkling family support. Flag the context conflict honestly — vendor model card says up to 1M; Artificial Analysis lists 256K in its article blurb — and use Index 40 as the independent baseline rather than vendor-claimed SWE-bench Verified 80.2% alone. Source: [Thinking Machines Lab; Artificial Analysis (independent Index 40); VentureBeat](https://thinkingmachines.ai/news/inkling-small/). ### MiniMax H3 **MiniMax** · 2026-07-31 · specialist · multimodal _Multimodal video model launches API-first with open weights promised in coming days — still not on Hugging Face or GitHub as of Aug 1_ Treat current state as API-only until the repo appears: unified text/image/video/audio understanding to video with native stereo, up to 15s at 2K, with vendor claims that 2K per-second price is under one-third of mainstream models. Do not build a self-host plan around the promise; the W30 K3 lesson applies again. Source: [MiniMax blog; SCMP; secondary weights-not-yet status](https://www.minimax.io/blog/minimax-h3). ## Architecture watch ### Post-training and harness policy move scores without new weights _Examples:_ DeepSeek V4-Flash-0731: +10 AA Index on unchanged 284B/13B, OpenAI ARC-AGI-3: 13.3% to 38.3% via retained reasoning + compaction, OpenAI Jul 29 serving/harness efficiency post (~20% serving cost, >15% token-gen efficiency). Two independent vendors moved measured capability this week without shipping new parameter counts: DeepSeek re-post-trained Flash in place, and OpenAI changed Responses API memory settings on the same Sol weights. Architects should treat reasoning retention, context compaction, and post-training checkpoints as first-class evaluation dimensions alongside the model ID. Buyers who only diff weight releases will miss the largest week-over-week gains. Source: [DeepSeek; Artificial Analysis; OpenAI](https://artificialanalysis.ai/articles/deepseek-v4-flash-0731-scores-50-on-the-artificial-analysis-intelligence-index-10-points-above-previous-deepseek-v4-flash). ### Revenue-tiered open licenses turn legal review into the deployment gate _Examples:_ Kimi K3 License: >$20M trailing-12-month MaaS commercial trigger, Kimi K3 attribution triggers above 100M MAU or $20M monthly revenue, Inkling-Small: clean Apache 2.0 contrast case. Downloadable weights no longer imply a single open-source diligence checklist. K3 is MIT-like until commercial hosting crosses the revenue trigger, at which point counsel must negotiate; Inkling-Small remains Apache 2.0 with no such trigger. Procurement should classify each open release by license economics before sizing clusters, and keep Apache-clean alternatives on the shortlist when MaaS revenue is material. Source: [MoonshotAI/Kimi-K3 LICENSE; Thinking Machines Lab](https://raw.githubusercontent.com/MoonshotAI/Kimi-K3/main/LICENSE). ### Serving stacks ship in the same week as the open MoE class _Examples:_ vLLM v0.26.0: Inkling modeling stack, NVFP4, MTP speculative decoding, Inkling-Small BF16 ~600 GB / NVFP4 ~180 GB VRAM floors, DeepSeek-V4 routing and speculative optimizations across NVIDIA/AMD/XPU. Production self-hosters got a same-week inference path for the new open MoE class rather than a multi-month serving lag. vLLM v0.26.0 (Jul 27) adds Inkling CUDA-graph and Hopper FA4 relative attention support alongside DeepSeek-V4 serving pushes and matured KV offloading. Capacity planners should price FP4 datapaths and KV-tiered storage into the model-release decision, not as a later optimization. Source: [vLLM project GitHub release v0.26.0; Thinking Machines model card](https://github.com/vllm-project/vllm/releases/tag/v0.26.0). ### Near-iso Intelligence at a fraction of active parameters _Examples:_ Inkling-Small: AA Index 40 vs Inkling 41 at <1/3 parameters, Inkling-Small: 12B active vs Inkling 41B active, DeepSeek V4 Flash prior: also sat at AA Index 40 in the same size class. Artificial Analysis places Inkling-Small within one Index point of full Inkling while activating 12B rather than 41B parameters, and notes no smaller open-weights model scores higher in that size class. Efficiency MoEs are now a procurement tier of their own: teams should ask whether the last Index point is worth 3x the serving footprint before defaulting to the flagship open checkpoint. Source: [Artificial Analysis (independent)](https://artificialanalysis.ai/articles/inkling-small-lands-within-a-point-of-inkling-on-the-artificial-analysis-intelligence-index-with-less-than-a-third-of-the-parameters). ## Benchmark moves ### Artificial Analysis Intelligence Index — DeepSeek V4 Flash Independent Index jumps from 40 to 50 (+10) for Flash-0731 on unchanged architecture; now ~1 point behind GPT-5.6 Luna (51) and 6 points above V4 Pro on AA's comparison - DeepSeek V4 Flash 0731 (max): 50 Index; GDPval-AA v2 1559 Elo; Terminal-Bench 2.1 79% (AA) - Prior DeepSeek V4 Flash: 40 Index; GDPval-AA v2 1189 Elo - Kimi K3 (max): 57 — open-frontier reference - GPT-5.6 Luna (max): 51 - GLM-5.2 (max): 51 Source: [Artificial Analysis (independent)](https://artificialanalysis.ai/articles/deepseek-v4-flash-0731-scores-50-on-the-artificial-analysis-intelligence-index-10-points-above-previous-deepseek-v4-flash). ### Artificial Analysis Intelligence Index — Inkling-Small New open US model lands at Index 40, within 1 point of Inkling (41) at less than a third the parameters - Inkling-Small: 40 (independent) - Inkling: 41 (independent) - DeepSeek V4 Flash (prior max, same size class): 40 - MiniMax-M2.7 (per AA): ~38 — about 2 points below Inkling-Small Source: [Artificial Analysis (independent)](https://artificialanalysis.ai/articles/inkling-small-lands-within-a-point-of-inkling-on-the-artificial-analysis-intelligence-index-with-less-than-a-third-of-the-parameters). ### ARC-AGI-3 public set (vendor harness, not official board) OpenAI reports GPT-5.6 Sol rising from 13.3% to 38.3% by retaining reasoning and compacting context — same model, roughly 6x fewer output tokens - GPT-5.6 Sol + retained reasoning + compaction: 38.3% (vendor-reported) - GPT-5.6 Sol + rolling truncation baseline: 13.3% (vendor-reported) Source: [OpenAI (vendor-reported; not an official ARC leaderboard reprint)](https://openai.com/index/how-two-settings-tripled-our-arc-agi-3-scores/). ### DeepSeek Terminal-Bench 2.1 (vendor harness vs independent) Vendor-claimed max-effort 82.7 versus Artificial Analysis 79% on the same Flash-0731 checkpoint — pin the harness before ranking - DeepSeek V4 Flash 0731 (DeepSeek harness, max effort): Terminal-Bench 2.1 82.7 (vendor-claimed) - DeepSeek V4 Flash 0731 (Artificial Analysis): Terminal-Bench 2.1 79% (independent) - DeepSeek V4 Flash 0731 (DeepSeek harness): Cybergym 76.7; Toolathlon verified 70.3; DeepSWE 54.4 (vendor-claimed) Source: [DeepSeek API changelog (vendor); Artificial Analysis (independent)](https://api-docs.deepseek.com/updates/). ## Tier scorecard _As of 2026-08-01._ | Tier | Leader | Challenger | Note | |---|---|---|---| | Closed frontier | Claude Fable 5 | Claude Opus 5 | No new closed flagship shipped; Fable retains the absolute reference while Opus 5 remains the economic default at $5/$25. | | Open frontier | Kimi K3 | Inkling-Small | K3 takes the lead now that weights and license text are public (AA Index 57); Inkling-Small is the efficiency challenger at Index 40 under Apache 2.0. | | Reasoning | Claude Fable 5 | GPT-5.6 Sol | Sol's vendor-reported ARC-AGI-3 harness jump shows memory policy can move scores without new weights; independent cost-per-task still pending. | | Coding | Claude Opus 5 | DeepSeek V4 Flash 0731 | Opus 5 holds the practical coding lead on W30 launch evidence; Flash-0731 is the cheap-tier agent upgrade at Index 50 and unchanged $0.14/$0.28. | | Multimodal | Gemini 3.6 Flash | Kimi K3 | Gemini remains the public fleet multimodal default; K3 now adds self-host control with MoonViT-V2. Gemini 3.5 Pro still has not GA'd. | | Edge / small | Inkling-Small | Gemini 3.5 Flash-Lite | Inkling-Small brings open multimodal agents to ~180 GB NVFP4; Flash-Lite remains the hosted speed-cost floor at roughly 350 tok/s. | ## Vendor signals - **2026-07-30 · OpenAI — GPT-5.6 Luna cut 80% to $0.20/$1.20 and Terra 20% to $2/$12; Sol held at $5/$30; Fast mode replaces Priority Processing** - Closed labs are competing on completed-task economics via rate cards and latency tiers rather than new flagship weights this week. Route only latency-sensitive work into Fast mode and keep ordinary asynchronous agents on standard Luna or Terra after measuring completion quality. - _Source:_ [OpenAI](https://openai.com/index/advancing-the-price-performance-frontier-with-gpt-5-6/) - **2026-07-27 · Moonshot AI — Kimi K3 License ships with weights: MIT-like until MaaS revenue exceeds $20M trailing twelve months** - Open-weights diligence now splits into quality evaluation and license classification. Counsel should map expected MaaS and product revenue against the trigger before any production self-host commitment; Apache-clean alternatives remain available for teams that cannot accept the commercial clause. - _Source:_ [MoonshotAI/Kimi-K3 LICENSE; Hugging Face](https://raw.githubusercontent.com/MoonshotAI/Kimi-K3/main/LICENSE) - **2026-07-31 · Google DeepMind — Gemini 3.5 Pro still not generally available as the standing Jul 31 prediction deadline expires** - Do not plan Q3 procurement around 3.5 Pro. Buy the callable Flash and Flash-Lite fleet now; another slip further weakens roadmap-based purchasing. - _Source:_ [Google (Jul 21 partner-testing status; no in-window GA found)](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-6-flash-3-5-flash-lite-3-5-flash-cyber/) - **2026-07-31 · MiniMax — H3 launches with an explicit promise to open weights in the coming days; no public HF or GitHub artifacts as of Aug 1** - Repeat the K3 lesson: an open-weights commitment is not a downloadable artifact. Keep H3 on the API watchlist until the repo and license appear. - _Source:_ [MiniMax blog; secondary weights-not-yet reports](https://www.minimax.io/blog/minimax-h3) - **2026-07-27 · vLLM project — vLLM v0.26.0 ships Inkling family support, NVFP4, MTP speculative decoding, and DeepSeek-V4 serving optimizations** - Serving readiness is now part of the model-release race. Self-host teams evaluating Inkling-Small or DeepSeek V4 this week should pin v0.26.0 as the baseline inference path rather than waiting for a later stack catch-up. - _Source:_ [vLLM GitHub releases](https://github.com/vllm-project/vllm/releases/tag/v0.26.0) ## Watchlist - **Next 7-14 days — MiniMax H3 downloadable weights and license.** Decides whether H3 becomes a real open video-generation option or remains an API product wrapped in open-language marketing. - **Next 30 days — DeepSeek V4-Pro official GA and pricing.** Flash-0731 is not Pro GA; the standing surge-pricing and official Pro triggers remain incomplete until a callable Pro ID and rate card land. - **Next 14 days — Inkling-Small context window: vendor 1M vs AA 256K.** Procurement specs need a resolved context number; watch for an AA methodology note or Thinking Machines clarification. - **By Aug 15 — Independent Opus 5 cost-per-task results.** Vendor benchmark leadership from W30 is still not enough; the market needs task cost, token use, and latency against Fable 5 and GPT-5.6 Sol. - **Next 2-3 weeks — LMArena Opus 5 Elo stabilization.** No qualifying in-window official board move; secondary pages remain unstable — wait for a date-stamped stable print before re-ranking coding. - **Next 30 days — Gemini 3.5 Pro callable model ID and public pricing.** The Jul 31 miss is now a track-record item; only a public GA closes the W30 prediction cleanly. ## Changelog - W31 opens on the Kimi K3 weights-and-license conversion, Inkling-Small Apache 2.0 efficiency play, DeepSeek Flash-0731 independent +10 Index jump, and OpenAI Luna/Terra cuts plus harness ARC-AGI-3 result. - Negative findings recorded as content: Gemini 3.5 Pro missed Jul 31 GA; no qualifying LMArena/SWE-bench/ARC official leaderboard moves; MiniMax H3 weights promised but not shipped. - Tree delta lists kimi-k3 under added for this issue's openness conversion while models.yaml still awaits an Inkling-Small row; editor should restructure to added:[inkling-small] / updated:[kimi-k3] once the YAML lands. --- Source of truth: `src/data/industry/models/2026-W31.ts`. Canonical HTML: . PDF: . Tree: .