DeepSeek V4.1 Flash lands on Merge Gateway, tops Terminal-Bench 2.1 at 90.6
shensi · x · 2026-09-12
Third-party API gateway Merge has onboarded DeepSeek V4.1 Flash. DeepSeek reports a leap in intelligence, with the model scoring 90.6 on Terminal-Bench 2.1 and 88.1 on CyberGym — the top scores in its comparison set.
Per Merge's model catalog, V4.1 Flash offers 1M context and 384K max output, priced at $0.15/M input (peak $0.30/M) and $0.60/M output (peak $1.20/M), keeping DeepSeek's aggressive pricing. The catalog also lists competing models from other vendors.
More from Models
- GPT-Image-2.5 aces complex prompts but falls short of Midjourney's aesthetic sensibility — perilli · 2026-09-12
- OpenAI releases GPT-6 Astra video: "needs less babysitting" — OpenAI · 2026-09-12
- Draft model hits ~60 tok/s running Qwen3.8-27B at 131k context on a 16GB GPU — pneuny · 2026-09-12
- TRACES Board: GPT-6-astra Tops Agent Leaderboard but Trails Opus-5 on Self-Repair — thisdudelikesAI · 2026-09-12
- DeepSeek v4.1 Flash on-device test: q2 runs at 16 tok/s but tool calls go off the rails — challis88ocarina · 2026-09-12
- Claude 5x Pro User Claims Post-Reset Quota Burns Far Faster Than Day One — unknown9645 · 2026-09-12