The week across the frontier labs — new defaults, previews and open-weight releases — plus where the coding benchmarks stand.
Claude Opus 5 — Released 24 Jul, Claude Opus 5 is Anthropic's fourth Claude 5 model in less than two months. Priced identically to Opus 4.8 at $5/$25 per million input/output tokens, the model is positioned to deliver near-Fable 5 performance at half the cost, and features a user-controllable effort toggle (low/medium/high). It replaces Opus 4.8 as the default on Claude Max and the strongest model on Claude Pro, with two new beta API features: mid-conversation tool changes and automatic safety-classifier fallbacks. · 24 Jul
Read → Axios · InfoWorld · FortuneMCP Spec Update (2026-07-28) — On 28 Jul Anthropic published a hardened MCP spec refresh adding a stateless core, standardized extensions, and tightened auth. The update is rolling out across Claude products and targets developer friction in bringing production MCP servers to Claude's connectors directory. Anthropic described the move as a continued investment in MCP as an open standard alongside the community. · 28 Jul
Read → Releasebot / Anthropic release notesKimi K3 — Moonshot AI dropped Kimi K3's full weights on Hugging Face on 26 Jul (a day ahead of schedule), published as 96 shards totalling ~1.56 TB under a custom Kimi K3 License. The model is a sparse MoE with 2.8 trillion total parameters (104B active per token), a 1M-token context window, and native multimodal input. API pricing is $3.00/$15.00 per million tokens; Together AI and Modal launched day-0 hosted inference. Anthropic CEO Dario Amodei responded publicly on 27 Jul, calling non-dangerous open models 'a public good' while still advocating chip export controls. · 26 Jul
Read → felloai.com · TechTimes · explainx.aiGrok STT 1.0 — xAI (operating as SpaceXAI) released Grok STT 1.0 on 23 Jul 2026, a speech-to-text model added to LLM Gateway on 24 Jul. The release expands SpaceXAI's model portfolio beyond text and coding, broadening the Grok platform toward multimodal API parity with OpenAI's Whisper and Google's Speech services. · 23 Jul
Read → LLM Gateway timelineGemini 3.6 Flash / 3.5 Flash Lite — Google released both Gemini 3.6 Flash ($1.50/$7.50 per million tokens, reported 17% more token-efficient) and Gemini 3.5 Flash Lite on 21 Jul 2026; both were added to LLM Gateway on the same day. The dual release targets cost-sensitive enterprise and consumer workloads as Google Cloud's contract backlog neared $460 billion. · 21 Jul
Read → LLM Gateway timeline · AIToolsRecapGrok Build (Apache 2.0) — Following a repo-upload privacy incident, SpaceXAI open-sourced Grok Build under Apache 2.0 on or around 25 Jul, releasing its coding agent harness to developers. The harness is compatible with any underlying model including Claude, widening the agentic tooling ecosystem beyond SpaceXAI's own models. · 25 Jul
Read → AIToolsRecapThe full AI Models Comparison dashboard — every category, factor weights, best-for and sources. Benchmark-anchored; ✅ verified · ⚠️ partial · ❓ released, not yet independently benchmarked. Wide tables scroll sideways.
| # | Model | Score | Business use | Web research | Presentation | Low halluc. | Coding | Tool use | Docs | Cost (in/out $) | Open? | Conf. |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5 | 93 | ● 95 | ● 90 | ● 93 | ● 96 | ● 97 | ● 95 | ● 94 | $10 / $50 | No | ✅ |
| 2 | Claude Opus 4.8 | 90 | ● 93 | ● 82 | ● 91 | ● 92 | ● 93 | ● 92 | ● 93 | $5 / $25 | No | ✅ |
| 3 | GPT-5.6 Sol | 89 | ● 90 | ● 90 | ● 88 | ● 82 | ● 92 | ● 92 | ● 90 | $5 / $30 | No | ✅ |
| 4 | Claude Sonnet 5 | 86 | ● 90 | ● 80 | ● 88 | ● 89 | ● 89 | ● 90 | ● 90 | $3 / $15* | No | ✅ |
| 5 | Kimi K2.6 | 86 | ● 88 | ● 92 | ● 88 | ● 95 | ● 90 | ● 94 | ● 88 | $0.95 / $4.00 | Yes | ✅ |
| 6 | GLM-5.2 | 83 | ● 78 | ● 78 | ● 76 | ● 80 | ● 88 | ● 86 | ● 76 | $1.40 / $4.40 | Yes | ✅ |
| 7 | Qwen 3.7-Max | 83 | ● 80 | ● 74 | ● 78 | ● 78 | ● 86 | ● 88 | ● 76 | $1.25 / $3.75* | No | ✅ |
| 8 | GPT-5.6 Terra | 82 | ● 84 | ● 84 | ● 80 | ● 80 | ● 86 | ● 85 | ● 84 | $2.50 / $15.00 | No | ✅ |
| 9 | Grok 4.5 | 82 | ● 78 | ● 82 | ● 76 | ● 78 | ● 86 | ● 84 | ● 76 | $2 / $6 | No | ✅ |
| 10 | Kimi K2.7-Code | 82 | ● 78 | ● 80 | ● 76 | ● 90 | ● 93 | ● 90 | ● 78 | $0.95 / $4.00 | Yes | ⚠️ |
| 11 | GLM-5V-Turbo (vision) | 81 | ● 78 | ● 76 | ● 82 | ● 76 | ● 90 | ● 86 | ● 82 | $1.20 / $4.00 | No | ⚠️ |
| 12 | Gemini 3.5 Flash | 81 | ● 76 | ● 88 | ● 78 | ● 80 | ● 85 | ● 82 | ● 76 | $1.50 / $9.00 | No | ✅ |
| 13 | DeepSeek V4-Pro | 81 | ● 78 | ● 75 | ● 76 | ● 82 | ● 86 | ● 80 | ● 78 | $0.44 / $0.87 | Yes | ⚠️ |
| 14 | MiniMax M3 | 78 | ● 75 | ● 88 | ● 76 | ● 75 | ● 82 | ● 80 | ● 78 | $0.30 / $1.20 | Yes | ⚠️ |
| 15 | DeepSeek V4-Flash | 75 | ● 70 | ● 68 | ● 68 | ● 74 | ● 82 | ● 74 | ● 70 | $0.14 / $0.28 | Yes | ✅ |
| 16 | MiMo-V2.5-Pro | 73 | ● 72 | ● 70 | ● 72 | ● 74 | ● 80 | ● 76 | ● 74 | $0.44 / $0.87 | No | ✅ |
| 17 | GPT-5.6 Luna | 72 | ● 72 | ● 74 | ● 70 | ● 76 | ● 74 | ● 74 | ● 72 | $1 / $6 | No | ✅ |
| 18 | ERNIE 5.1 | 72 | ● 74 | ● 72 | ● 72 | ● 74 | ● 74 | ● 72 | ● 74 | $0.59 / $2.65 | No | ✅ |
| 19 | Qwen 3.7-Plus | 72 | ● 74 | ● 72 | ● 74 | ● 76 | ● 82 | ● 76 | ● 74 | $0.32 / $1.28 | No | ⚠️ |
| # | Model | Score | Egyptian quality | Naturalness | Emotion | Cloning | Real-time | $/min audio | Conf. |
|---|---|---|---|---|---|---|---|---|---|
| 1 | gpt-realtime-1.5 | 90 | ● 95 | ● 92 | ● 85 | ● 70 | ● 85 | ~$0.18–0.46/min | ⚠️ |
| 2 | Grok Voice Agent | 88 | ● 95 | ● 90 | ● 85 | ● 88 | ● 95 | $0.05/min · 2-min clone | ✅ |
| 3 | gpt-realtime-mini | 80 | ● 92 | ● 85 | ● 85 | ● 60 | ● 90 | $0.063–0.146/min | ❓ |
| 4 | Gemini 3.1 Flash Live | 78 | ● 82 | ● 86 | ● 78 | ● 0 | ● 92 | ~$0.008/min | ⚠️ |
| 5 | GLM-4-Voice | 55 | ● 30 | ● 80 | ● 75 | ● 60 | ● 90 | Open · CN/EN only | ⚠️ |
| # | Model | Score | Egyptian quality | Naturalness | Emotion | Cloning | Latency | $/min audio | Conf. |
|---|---|---|---|---|---|---|---|---|---|
| 1 | ElevenLabs v3 Multilingual | 88 | ● 92 | ● 95 | ● 95 | ● 95 | ● 80 | $0.10–0.30/1K chars · Egyptian ✅ | ✅ |
| 2 | Grok TTS | 86 | ● 90 | ● 88 | ● 80 | ● 85 | ● 92 | $4.20/1M chars · 120s clone | ✅ |
| 3 | ElevenLabs Flash / Turbo | 78 | ● 85 | ● 88 | ● 80 | ● 90 | ● 95 | $0.05/1K chars · ~75ms | ✅ |
| 4 | Microsoft MAI-Voice-2 | 70 | ● 60 | ● 85 | ● 78 | ● 0 | ● 80 | $22/1M chars · Arabic ⚠️ | ⚠️ |
| 5 | GLM-TTS | 68 | ● 55 | ● 85 | ● 80 | ● 88 | ● 85 | Open · 3s clone | ⚠️ |
| 6 | Mistral Voxtral TTS | 66 | ● 55 | ● 82 | ● 72 | ● 0 | ● 80 | $0.016/1K chars | ⚠️ |
| # | Model | Score | Egyptian / Arabic | Languages | Real-time | Accuracy | $/min audio | Conf. |
|---|---|---|---|---|---|---|---|---|
| 1 | Grok STT | 82 | ● 85 | 25 | ● 90 | ● 92 | $0.10/hr batch · diarization | ✅ |
| 2 | GLM-ASR-2512 | 76 | ● 80 | 30 | ● 90 | ● 88 | Open · Arabic ✅ | ⚠️ |
| 3 | ElevenLabs Scribe v2 | 76 | ● 78 | 90 | ● 90 | ● 91 | $0.39/hr · 150ms | ❓ |
| 4 | OpenAI gpt-4o-transcribe | 74 | ● 70 | 99 | ● 85 | ● 90 | $0.003/min mini | ⚠️ |
| 5 | gpt-realtime-whisper | 72 | ● 70 | 99 | ● 92 | ● 88 | $0.017/min streaming | ⚠️ |
| 6 | Gemini 3.1 Flash STT | 70 | ● 68 | 90 | ● 88 | ● 86 | ~$0.005–0.01/min | ⚠️ |
| # | Model | Maker | Score | Arena Elo | Text in image | Arabic text | Prompt adh. | Speed | $/image | Open? | Conf. |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | GPT Image 2 (high) | OpenAI | 93 | 1339 | ● 99 | ● 95 | ● 92 | ● 75 | $0.211 (high) · $0.053 (med) | No | ✅ |
| 2 | Reve 2.0 | Reve | 84 | 1281 | ● 88 | ● 60 | ● 88 | ● 82 | $0.024 · $0.04 edit | No | ⚠️ |
| 3 | MAI-Image-2.5 | Microsoft | 83 | 1271 | ● 85 | ● 60 | ● 86 | ● 80 | ~$0.02–0.06 (token) | No | ⚠️ |
| 4 | GPT Image 1.5 (high) | OpenAI | 82 | 1260 | ● 90 | ● 82 | ● 86 | ● 75 | $0.16 (high) | No | ✅ |
| 5 | Qwen-Image 2.0 | Alibaba | 82 | 1180 | ● 92 | ● 88 | ● 88 | ● 78 | ~$0.05 (varies) | Yes | ✅ |
| 6 | HiDream-O1-Image 1.5 | HiDream | 81 | 1265 | ● 84 | ● 58 | ● 88 | ● 80 | credit-based | Yes | ⚠️ |
| 7 | Cosmos3-Super | NVIDIA | 80 | 1227 | ● 78 | ● 55 | ● 80 | ● 78 | Free (open, self-host) | Yes | ⚠️ |
| 8 | Gemini 3.1 Flash Image | 80 | 1170 | ● 85 | ● 70 | ● 86 | ● 90 | $0.067 (1K) · $0.101 (2K) | No | ✅ | |
| 9 | HunyuanImage 3.0 | Tencent | 80 | 1152 | ● 88 | ● 85 | ● 88 | ● 80 | Free (open 80B MoE) | Yes | ✅ |
| 10 | Ideogram 4.0 | Ideogram | 79 | 1200 | ● 96 | ● 60 | ● 90 | ● 82 | $0.03 · $0.06 · $0.10 | Partial | ⚠️ |
| 11 | Imagen 4 Ultra | 79 | 1150 | ● 80 | ● 68 | ● 88 | ● 60 | $0.06 · $0.02 Fast | No | ⚠️ | |
| 12 | Seedream 5.0 Lite | ByteDance | 78 | 1150 | ● 80 | ● 58 | ● 84 | ● 88 | $0.031 + $0.007 web | No | ⚠️ |
| 13 | CogView-4 | Zhipu | 78 | 1180 | ● 82 | ● 45 | ● 88 | ● 80 | $0.03 · Apache 2.0 | Yes | ⚠️ |
| 14 | FLUX.2 Pro | Black Forest Labs | 77 | 1140 | ● 80 | ● 55 | ● 86 | ● 85 | $0.03–0.073 | Partial | ✅ |
| 15 | Grok Imagine 1.5 | xAI | 72 | 1173 | ● 70 | ● 55 | ● 78 | ● 85 | $0.02 | No | ❓ |
| 16 | Luma Photon | Luma Labs | 70 | 1100 | ● 72 | ● 50 | ● 82 | ● 92 | $0.015 · $0.002 Flash | No | ❓ |
| # | Model | Maker | Score | Max resolution | Max duration | $/clip | Open? | Conf. |
|---|---|---|---|---|---|---|---|---|
| 1 | CogVideoX-1.5 | Zhipu | 72 | 768×1360 | ~10s @ 16fps | Open (self-host) | Yes | ⚠️ |
| 2 | CogVideoX | Zhipu | 70 | 1440×960 | ~30s | Open (self-host) | Yes | ✅ |
| 3 | Ying (Qingying) | Zhipu | 68 | 1440×960 | ~6s clips | CN market · n/a | No | ⚠️ |
| Factor | Weight % | Tier |
|---|---|---|
| Hallucination level (low = better) | 16 | critical |
| Following instructions exactly | 11 | high |
| Document generation quality | 9 | high |
| Long task stamina | 9 | high |
| Tool use / agent reliability | 8 | high |
| Coding quality — real software | 7 | medium |
| Skills / tools per conversation | 7 | medium |
| Research & web browsing | 7 | medium |
| Hard reasoning / problem solving | 5 | medium |
| Image / chart understanding | 5 | medium |
| Cost — input + output | 4 | lower |
| Knowledge depth (PhD-level) | 4 | lower |
| Frontend / web design | 3 | lower |
| Coding — long projects | 3 | lower |
| Math accuracy | 1 | lower |
| Memory — claimed max | 1 | lower |
| Factor | Weight % | Tier |
|---|---|---|
| How natural it sounds (Egyptian) | 22 | critical |
| Pronunciation accuracy (Egyptian) | 18 | critical |
| Emotion & expressiveness | 12 | high |
| Voice cloning quality | 12 | high |
| Real-time conversation (S2S) | 10 | high |
| Steerability (tone control) | 8 | medium |
| Speed to first word | 7 | medium |
| Speech-to-text included | 5 | medium |
| Cost | 4 | lower |
| Voice variety | 2 | lower |
AI leaders and their capital cycle — capex commitments alongside the month’s market moves.
(27 Jul) An analyst maintained an Overweight rating on NVDA and raised the price target from $310 to $330, as Nvidia reported July market cap around $4.83T; the stock's 52-week range is $164.07–$236.54 with YTD gain of roughly 8.3% per Yahoo Finance data as of 27 Jul.
Read full article → Yahoo Finance · Robinhood(30 Apr / Q1 earnings) Alphabet lifted its full-year 2026 capex guidance by $5 billion to as much as $190 billion, matching Microsoft; Google Cloud's contract backlog nearly doubled to ~$460 billion, driven by enterprise AI and TPU hardware agreements. Shares climbed ~7% after hours.
Read full article → Yahoo Finance · Tom's Hardware(30 Apr / Q1 earnings) Meta increased its full-year 2026 capex projection by $10 billion to a range topping $145 billion, citing higher memory component pricing and growing competition for land, power, and skilled workers. Revenue grew 33% to $56.3 billion in Q1.
Read full article → Tom's Hardware · Yahoo Finance(25 Jul) AIToolsRecap reported that Kimi K3's open-weight release was associated with ~$314 billion in pre-IPO valuation declines across Anthropic (~$232B) and OpenAI (~$82B) on secondary markets, as Moonshot AI temporarily suspended new subscriptions due to capacity constraints following the launch.
Read full article → AIToolsRecap(May 2026) Anthropic closed a $65 billion Series H led by Altimeter Capital, Dragoneer Investment Group, Greenoaks, and Sequoia Capital, pushing its post-money valuation to $965 billion and surpassing rival OpenAI's $852 billion mark. The raise comes just three months after Anthropic's $30 billion Series G at $380 billion, representing a roughly 22× valuation increase in 18 months. Compute partners include Amazon (up to 5GW AWS), Google/Broadcom (5GW TPU), and SpaceX Colossus.
Read full article → Morningstar / PitchBook(March 2026) OpenAI closed a $122 billion round — the single largest private venture round in history — pushing its post-money valuation to $852 billion. Amazon was named exclusive third-party cloud partner as part of a $50 billion cloud commitment. OpenAI reported 900 million weekly active users and over $20 billion in annualised revenue, and is preparing for a potential IPO targeting a near-$1 trillion valuation in Q4 2026.
Read full article → Crescendo.ai(May 2026; confirmed in SpaceX S-1) Anthropic signed a deal to access all compute capacity at SpaceX's Colossus 1 data centre in Memphis (300MW, ~220,000 Nvidia GPUs) at a reported $1.25 billion per month through May 2029 — approximately $15 billion per year. The agreement doubled Claude Code rate limits and removed peak-hour restrictions for Pro/Max users. The deal also includes an expression of interest in multiple gigawatts of space-based compute development.
Read full article → CNBC · Latent Space / AINews · Enterprise DNA(22 Jun 2026) Open-source AI startup Reflection AI (last valued at $25 billion) signed a compute deal with SpaceX for access to Colossus infrastructure starting 1 Jul 2026, paying $150 million per month — totalling roughly $6.3 billion if the contract runs through 2029. The deal includes immediate access to Nvidia GB300 accelerators and illustrates SpaceX's rapid commercialisation of Colossus as a third-party compute platform after landing Anthropic, Google, and Cursor as customers.
Read full article → CNBC(Apr–Jun 2026 earnings) Amazon (~$200B), Microsoft (~$190B), Alphabet ($180–190B), and Meta ($115–145B) collectively guided to approximately $725 billion in 2026 capital expenditure, up 77% from ~$410 billion in 2025. Goldman Sachs now projects a combined $5.3 trillion in capex for these four hyperscalers from FY2025 to FY2030. Analysts forecast combined hyperscaler capex will surpass $1 trillion in 2027.
Read full article → Yahoo Finance · ValueAdd VC · Goldman Sachs via Yahoo Finance(1 Jul 2026) Together AI closed an $800 million Series C led by Aramco Ventures, with Nvidia, Vista Equity Partners, General Catalyst, Emergence, Salesforce Ventures, and others participating, at a post-money valuation of ~$8.3 billion. Annual bookings had already crossed $1.15 billion in Q2 2026. The round signals sustained institutional appetite for GPU-cloud and open-model inference platforms.
Read full article → TechCrunch via VentureCapitalTracker · TechStartups(Jul 2026) Abu Dhabi's AI investment firm MGX closed Fund I at $49 billion, exceeding its $45 billion target. The fund has invested in 14 companies across semiconductors, AI infrastructure, and AI platforms. MGX is developing Europe's largest AI campus near Paris targeting 3GW of compute capacity and acquired Aligned Data Centres as part of a $40 billion consortium transaction.
Read full article → Crescendo.ai · VentureCapitalTracker(27 Jul 2026) Gritt Robotics emerged from stealth with a $26 million Series A led by First Round Capital, Union Square Ventures, Congruent Ventures, and Active Impact Investments. The company builds AI-powered robotic systems integrated with construction hardware, claiming robots that quadruple solar panel installation rates — a direct play on AI data-centre power buildout driving grid-scale renewable demand.
Read full article → Parsers Substack weekly funding report (28 Jul) · Bricks & Bytes