HOME › REVIEWS › AI MODELS

82 reviews. AI Models.

AVG SCORE: 6.9/10

82 of 1403 reviews AI Models × Clear all

"Sonnet 4.6 delivers Opus-level autonomy in agentic coding but burns tokens 15-45% faster, raising costs for both subs…"

AI Models 47 sources · <60%

"Users report real speed gains and workflow wins, but the community sees incremental advances buried under hype and fo…"

AI Models 150 sources · 90%+

"Capable frontier model with strong intent inference and generous sub limits, but HN devs report 'austere' code output…"

AI Models 63 sources · <60%

"Frontier LLM praised by video reviewers as a daily workhorse but flagged by developers for high API costs and surface…"

AI Models 26 sources · <60%

"Influencers call it a 'workhorse,' but seasoned HN developers question whether AI-generated code holds up in producti…"

AI Models 26 sources · <60%

"A mini-distilled variant of GPT-5.6 positioned as a cost-efficient coding and agent-orchestration model — smaller tha…"

AI Models 113 sources · 90%+

"Tencent's Hy3 is an open-source LLM praised for aggressive pricing but criticized for benchmark mediocrity and incons…"

AI Models 333 sources · 90%+

"Alibaba's open-weights LLM excels in local deployment and inference economics, but carries censorship, thought-loop, …"

AI Models 43 sources · <60%

"A highly cost-effective LLM with impressive multimodal potential, hindered by UI issues, hallucinations, and poor vid…"

AI Models 205 sources · 90%+

"Sonnet 5 shows polarizing utility for coding, with users debating skill atrophy while reviewers struggle with confusi…"

AI Models 137 sources · 90%+

"Analysis of Exotic AI- reveals a focus on local model hosting, high hardware costs, and significant concerns over AI-…"

AI Models 429 sources · 90%+

"A high-performance reasoning model excelling in math and coding via pure RL, though local inference is slow and censo…"

AI Models 78 sources · 60-89%

"Opus 4.8 ships with reported token-burning errors, slowness, and lazy thinking — users revert to 4.7/4.6. Scam-proxy …"

5.8

/10

24 JUN

Grok by SpaceXAI for Word

AI Models 29 sources · <60%

"Grok ranges from a Word add-in to a general LLM. Feedback is polarized between practical drafting utility and bizarre…"

AI Models 116 sources · 90%+

"Engineers report massive productivity gains using ChatGPT, but face severe burnout and ethical boycotts over OpenAI's…"

AI Models 91 sources · 60-89%

"Hyped as a frontier open-weight coding model, but user data reveals skewed adoption metrics and mixed real-world perf…"

AI Models 60 sources · 60-89%

"Strong multimodal and tool-calling performance, but users report session-to-session inconsistency. Competitive with K…"

AI Models 89 sources · 60-89%

"Moonshot's open-weights coding model offers strong value but trails Claude Opus in real-world coding quality per deve…"

AI Models 149 sources · 90%+

"Frontier AI model praised for complex coding but criticized for rate limits, cost, and aggressive guardrails. Real us…"

AI Models 203 sources · 90%+

"User discussions focus on AI philosophy and skepticism, while video tests show mixed results for coding and cost effi…"

2.1

/10

19 JUN