ARCHER Baselines - CoT-FSP - GSM8K: leaderboard
Direct runs on identical instances; shared normalization/scoring; one response/example; AIME complete set single-run exactmatch. CoT-FSP uses5released standardbackbones; Vanilla uses2released reasoningmodels. Fine-tuned CoT-FT/Toolformer/CoA/ARCHER rows require distinct trainedcheckpoint identity and are withheld. Qwen3-8B has thinking enabled. Metric: Final-answer accuracy (%). Source: https://arxiv.org/html/2611.18494v1
Metric: Final-answer accuracy (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 5 models tracked. Updated 17 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Qwen 3 8B | Thinking | 80.47 | #1868 |
| 2 | Qwen 3.5 4B | 77.63 | #1227 | |
| 3 | Falcon-H1R-7B | 73.61 | #1672 | |
| 4 | Gemma 4 E2B | 71.74 | #4291 | |
| 5 | Qwen 3 1.7B | 69 | #5807 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=archer-baselines-cot-fsp-gsm8k · How It Works · Data refreshed daily, snapshot 2026-11-17.