PulseAugur
EN
LIVE 21:28:17

Cosine Genie leverages GPT-4o fine-tuning to become top coding agent

By PulseAugur Editorial · [1 sources] ·

Cosine has launched Genie, a coding agent that has achieved the top ranking on the SWE-Bench benchmark, surpassing previous leaders by a significant margin. This success is attributed to fine-tuning OpenAI's GPT-4o model on billions of tokens of synthetically generated code and runtime errors. OpenAI collaborated with Cosine on the scale and specifics of the fine-tuning process, including the dynamic sizing of LoRA adapters. Genie utilizes a four-stage workflow and is designed to output code in formats suitable for direct integration into codebases. AI

RANK_REASON New coding agent (Genie) from Cosine achieves state-of-the-art results on SWE-Bench using fine-tuned GPT-4o, a significant advancement in AI coding capabilities.

Read on Latent Space Podcast →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Cosine Genie leverages GPT-4o fine-tuning to become top coding agent

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Frontier Release
New coding agent (Genie) from Cosine achieves state-of-the-art results on SWE-Bench using fine-tuned GPT-4o, a significant advancement in AI coding capabilities.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, product, paper
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
765 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [1]

  1. Latent Space Podcast TIER_1 English(EN) · Latent.Space ·

    Is finetuning GPT4o worth it? — with Alistair Pullen, Cosine (Genie)

    <p><a href="https://en.wikipedia.org/wiki/Betteridge%27s_law_of_headlines" target="_blank">Betteridge's law</a> says no: with seemingly infinite flavors of RAG, and >2million token context + prompt caching from Anthropic/Deepmind/Deepseek, it's reasonable to believe that "in cont…

RELATED ENTITIES

RELATED TOPICS