OpenAI's Yann Dubois says GPT-5.6-Sol beats Claude-Fable-5 on DeepSWE benchmark with lower token costs
GPT-5.6-Sol scored 72% at $8.39 per task.
Combined views
491.9K
13 Sources, first seen ago
Sources
AA 
Anastasios Nikolas Angelopoulos@ml_angelopoulosCheck it out! It's a Fable-quality model! https://twitter.com/arena/status/2075672492312768683
- likes: 13
- replies: 2
- bookmarks: 1
- reposts: 0
SW Sherwin Wu@sherwinwu- likes: 91
- replies: 2
- bookmarks: 8
- reposts: 6
RP Rohan Paul@rohanpaul_aiGPT 5.6 Beats Fable 5 by on DeepSWE at a cheaper price. GPT-5.6 Sol reaches roughly 72% to 73% DeepSWE at about $8.4 per task, while Claude/Fable-5 tops out around 70% at a much higher cost, around $13 to $22 per task. DeepSWE is a coding-agent benchmark, not a normal…
- likes: 83
- replies: 8
- bookmarks: 16
- reposts: 16
RP Rohan Paul@rohanpaul_aiToday’s edition of my newsletter just went out. 🔗 https://www.rohan-paul.com/p/gpt-56-beats-fable-5-by-on-deepswe 🗞️ GPT 5.6 Beats Fable 5 by on DeepSWE at a cheaper price. 🗞️ 1X Debuts Human-Level Robotics Hands - Neo’s Tendon-Driven Hands With 25 Degrees of Freedom 🗞️…
- likes: 29
- replies: 4
- bookmarks: 2
- reposts: 4
TT Theo - t3.gg@theogpt-5.6-sol is meaningfully better in Claude Code than in Codex I'm going to crash out so badly over this
- likes: 5.5K
- replies: 338
- bookmarks: 1K
- reposts: 64
WC Wei-Lin Chiang@infwinstonWhat a week - three frontier models (Grok-4.5, Muse-Spark-1.1, GPT-5.6) launched, and GPT-5.6 reached #1 in Code Arena, matching Claude Fable-5. The competition is heating up! https://twitter.com/arena/status/2075672492312768683
- likes: 60
- replies: 2
- bookmarks: 7
- reposts: 4