gpt-6.1-sol
PulseAugur coverage of gpt-6.1-sol — every cluster mentioning gpt-6.1-sol across labs, papers, and developer communities, ranked by signal.
21 day(s) with sentiment data
-
Open-weight AI model Darwin-180B-RSI outperforms larger rivals in reasoning tests
A new open-weight AI model named Darwin-180B-RSI, developed by Korea's VIDRAFT, has achieved a score of 73.68 in mathematical and scientific reasoning on the SuperCLUE evaluation. This performance places it second overa…
-
Coding agent workflow tested across Claude and Codex models
A developer conducted an experiment to test the effectiveness of a Scrum workflow for coding agents across seven different AI models, including Anthropic's Claude (Haiku, Sonnet, Opus 5.5) and models from Codex CLI (GPT…
-
LLM brand mention tracking faces measurement traps, study finds
A recent analysis explored the challenges of accurately measuring brand mentions in Large Language Model (LLM) responses, particularly when LLMs recommend products. The study focused on four web-search-enabled LLMs: Ope…
-
AI models struggle to accurately count fabric threads from smartphone photos
Researchers have developed a new method for measuring fabric thread density using smartphone photos taken through a pick glass, aiming for a $\pm$5% accuracy tolerance. They benchmarked several AI models, including Molm…
-
LLM token pricing explained: input, output, cache, and hidden costs
Understanding the cost of using large language models involves breaking down expenses by token type and provider-specific rates. Key cost factors include input tokens, output tokens, cached input, and cache writes, with…
-
AI code review system separates generation and verification roles
The author describes a system for using AI models to review code, separating the roles of code generation and code review. Claude Code is used for writing code, while Codex acts as a read-only reviewer to catch errors. …
-
Korean model Darwin debuts on SuperCLUE benchmark, ranks second in reasoning
The SuperCLUE benchmark, released in November 2026, highlighted the Darwin-180B-RSI-R3 model from Korea, placing it on the leaderboard for the first time. This model achieved second place in math and science reasoning a…
-
SuperCLUE benchmark ranks 19 top LLMs with monthly evolving private test sets · 6 sources tracked
The SuperCLUE benchmark has released its November 2026 rankings, featuring 19 advanced large language models. This benchmark is notable for its unique approach of redeveloping its private test set monthly, preventing mo…
-
Darwin-180B-RSI-R3 leads open-weight models on SuperCLUE math reasoning
A new open-weight model, Darwin-180B-RSI-R3, has achieved a score of 73.68 on SuperCLUE's mathematical and scientific reasoning benchmark, placing it first among open-weight models and second overall. This model, which …
-
AI retry budgets prevent overwhelming LLM providers during outages
A new approach to handling AI model retries aims to prevent overwhelming providers during outages. Instead of simply retrying failed calls multiple times, a "retry budget" limits the proportion of recent traffic that ca…
-
Developer's Nodast tool uses ASTs for LLM code edits, showing promise for structural changes
A developer has created a new method called Nodast for LLMs to edit code by manipulating abstract syntax trees (ASTs) rather than plain text. This approach uses S-expressions and a Hindley-Milner checker to validate edi…
-
New Frontier3DBench benchmarks AI coding agents for 3D creation
A new benchmark called Frontier3DBench has been introduced to evaluate coding agents' capabilities in 3D creation across objects, scenes, and interactive open worlds. The benchmark comprises 145 tasks and assesses agent…
-
VIDRAFT's Darwin-180B-RSI-R3 model listed as reference on SuperCLUE leaderboard
VIDRAFT's Darwin-180B-RSI-R3 large language model has been included as a reference model on the SuperCLUE November 2026 leaderboard. The model achieved a score of 43.08, placing it alongside other prominent frontier mod…
-
VIDRAFT's Darwin-180B-RSI-R3 claims #2 in global math/science reasoning
VIDRAFT has released Darwin-180B-RSI-R3, an open-weight Mixture-of-Experts large language model that utilizes Recursive Self-Improvement (RSI) to enhance its reasoning capabilities without new human-labeled data. The mo…
-
New CARTBench benchmark reveals LLM shopping advice gap
Researchers have introduced CARTBench, a new evaluation framework designed to assess the performance of large language models (LLMs) in providing shopping advice for complex and ambiguous requests, especially those invo…
-
AI agents in Cursor IDE exhibit unreliable behavior, frustrating users
Working with AI agents, particularly within the Cursor IDE using Composer 2.5, can be a frustrating experience due to their unpredictable behavior. Users report that agents may delete code, revert changes, or fail to co…
-
New Study Questions Multi-Estimate Debiasing for Language Models
A new arXiv paper investigates whether eliciting multiple weighted estimates from language models can uniformly debias their judgments. The study tested configurations of gpt-6.1-sol, Claude Sonnet 5.5, and Qwen3_8B acr…
-
Claude Code and Codex show equal command compatibility with minor format fixes
A practical test revealed that the same set of Chinese commands can be used effectively with both Claude Code and Codex, with only minor adjustments needed for compatibility. The study found that 10 out of 10 commands p…
-
Marketplace AI Unit Economics Formula Detailed
This article outlines a formula for calculating the unit economics of AI features in marketplaces, emphasizing the importance of defining the correct unit of analysis: the completed transaction. It details how to calcul…
-
Marketplaces build LLM cost observability to track AI spend
Marketplaces integrating AI features often face unexpected cost overruns due to a lack of detailed billing information. This article provides a practical guide to implementing LLM cost monitoring within a marketplace's …