Surgio-SWE Kimi K2.7 Edition
Specialized intelligence for real-world software engineering.
Better data. Bigger gains.
We train frontier models on our off-the-shelf data and put the resulting models to the test. Each training run demonstrates what great data makes possible: stronger benchmark scores, gains that generalize to new capabilities, and leaps in model behavior.
pass@134.8% → 56.1%pass@130% → 45%pass@118.9% → 28.9%pass@166.90% → 73.45%We trained Nemotron 3.5 Super on Surge data for agentic coding, long-horizon tool use, professional graphical reasoning, and instruction following.
pass@163.9% → 70.3%pass@186.9% → 92.1%pass@127.9% → 32.6%pass@114.8% → 27.5%With tools, pass@139.4% → 50.6%With tools, pass@179.35% → 85.80%With tools, pass@112.4% → 31.0%pass@154.8% → 67.6%pass@158.3% → 66.6%pass@127.3% → 35.5%pass@150.6% → 58.3%pass@172.3% → 76.8%Post-training GLM 5.3 on Frontier Work and GDPval-Verified data raised its GDPval-AA v2.1 Elo rating from 1650 to 1704 (+54 points). On DAYJOB, which remains quite difficult for GLM 5.3, the Healthcare task pass rate rose from 0% to 8%.
elo rating1650 → 1704Task Pass Rate0% → 8.0%Training on 2,400 expert-built coding tasks improved Kimi K2.7 across every external SWE eval we measured. This builds on an earlier run with 1,700 tasks, with continued performance improvements.
PASS@131.0% → 48.7%pass@160.1% → 68.1%pass@19.5% → 14.3%Post-training on tasks from the DAYJOB companion set raised Kimi K2.7's mean rubric score on AA-Briefcase-Lite by +8.4 pp.
Post-training GLM-5.3 on 750 tasks from the sudo L7 companion set improved pass@1 on DeepSWE by +4.8 pp and Terminal-Bench 4 by +6.6 pp.
Training on Surge coding data improved GDPval-AA from 30.7% to 37.6%, even though GDPval-AA contains no coding tasks.
Training on the DAYJOB companion set, which contains no coding tasks, nonetheless improved Terminal-Bench 2.1 by +8.4pp.
We post-trained Kimi K2.7 on 1,465 GDP.pdf companion tasks. The GDP.pdf holdout full-task pass rate rose +13.5pp and GDPval success more than doubled. The trained model used tools to find more of the relevant sources before creating deliverables, while reading less text overall.
Training on 1,700 Surge coding tasks improved Kimi K2.7 across every external SWE eval we measured. The biggest behavioral changes showed up in the complexity of real-world software engineering: carrying messy specifications through, testing beyond the obvious path, preserving existing behavior, and building its own checks when the environment did not provide one.
We trained on long-horizon office tasks with no coding examples. SWE-Bench Pro still improved +5.7pp. Trajectory analysis pointed to a broader capability we call goal-directed execution: choosing useful subgoals, maintaining task state, staying faithful to the objective, and verifying completion.
Every ComplexConstraints training example was single-turn, yet the largest gains appeared in multi-turn behaviors. Our hypothesis is that the underlying skill is shared: maintaining many interacting constraints without dropping them, whether they arrive in one dense prompt or accumulate over nine turns.
We trained Qwen3.5-122B-A10B on realistic, multi-tool agent environments. The gains transferred to Toolathlon, τ²-Bench, and BFCL-V4. Trajectory analysis showed better task closure and more efficient tool use across unfamiliar evaluation environments.
We trained GLM-4.7 on expert-authored STEM problems. It nearly doubled performance on held-out Expert STEM tasks, with gains also appearing on HLE, FrontierScience, PubMedQA, and DeepDive. The result suggests that difficult expert problem solving can teach reasoning patterns that remain useful across different scientific domains.
We trained GLM-4.6 inside CoreCraft, a high-fidelity customer-support environment. The gains transferred beyond CoreCraft to BFCL, τ²-Bench Retail, and Toolathlon. The result suggests that realistic enterprise workflows can teach general agent skills that survive changes in tools, tasks, and evaluation environments.
Meet our family of post-trained models. Each starts with an open foundation model and is trained on Surge AI data and environments.
The same off-the-shelf Surge AI datasets powering these runs are available to you. Train on the full dataset first, and pay only if it improves your metrics.