Benchmark Almanac

Next.js Agent Evals

Claude Fable 5.1 100% leads since September 2026
Not in model cards

Coding agents with their models do real Next.js tasks. The board reports success rate with and without an AGENTS.md file, plus duration and cost.

Rerun whenever the evals change, last on October 30th, 2026. Top agents reach 97%, so it is near saturation. About 330 GitHub stars.

Made by
Vercel
Released
October 2025
Kind
Web and app front ends
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Strongest model tested
GPT-6 Astra #3 on the capability index
Last model added
October 2026
Scores refreshed
November 12th, 2026
16
Entries on its own leaderboard, 14 models we track

5 models share the top score of 100%

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

16 entries. Scroll sideways for the rest.

  1. 100.0 Claude Fable 5.1
  2. 100.0 Claude Opus 5.5
  3. 100.0 GPT-6 Sol
  4. 100.0 Gemini 4 Argon
  5. 100.0 GPT-6.1 Terra
  6. 96.0 Grok 4.7
  7. 93.0 Pixel Canary
  8. 93.0 Gemini 3.8 Flash
  9. 93.0 GPT-6 Astra
  10. 86.0 Kimi K3
  11. 86.0 GPT-6 Luna
  12. 82.0 Cursor Composer 2.5
  13. 82.0 MiniMax-M3
  14. 82.0 GLM-5.2
  15. 82.0 Claude Sonnet 5
  16. 75.0 Kimi K2.7 Code
  1. 1 Claude Fable 5.1 high 100%
  2. 2 Claude Opus 5.5 high 100%
  3. 3 GPT-6 Sol high 100%
  4. 4 Gemini 4 Argon 100%
  5. 5 GPT-6.1 Terra high 100%
  6. 6 Grok 4.7 96%
  7. 7 Pixel Canary 93%
  8. 8 Gemini 3.8 Flash 93%
  9. 9 GPT-6 Astra high 93%
  10. 10 Kimi K3 86%
  11. 11 GPT-6 Luna high 86%
  12. 12 Cursor Composer 2.5 82%
  13. 13 MiniMax-M3 82%
  14. 14 GLM-5.2 82%
  15. 15 Claude Sonnet 5 82%
  16. 16 Kimi K2.7 Code 75%

Source: Official Next.js Agent Evals leaderboard

#EntryScoreSettingDateSource
1 Claude Fable 5.1 100% high, Claude Code Sep 2026 Official leaderboard
2 Claude Opus 5.5 100% high, Claude Code Sep 2026 Official leaderboard
3 GPT-6 Sol 100% high, Codex Sep 2026 Official leaderboard
4 Gemini 4 Argon 100% OpenCode Sep 2026 Official leaderboard
5 GPT-6.1 Terra 100% high, Codex Oct 2026 Official leaderboard
6 Grok 4.7 96% OpenCode Sep 2026 Official leaderboard
7 Pixel Canary 93% OpenCode Official leaderboard
8 Gemini 3.8 Flash 93% OpenCode Sep 2026 Official leaderboard
9 GPT-6 Astra 93% high, Codex Sep 2026 Official leaderboard
10 Kimi K3 86% OpenCode Jul 2026 Official leaderboard
11 GPT-6 Luna 86% high, Codex Sep 2026 Official leaderboard
12 Cursor Composer 2.5 82% Cursor Official leaderboard
13 MiniMax-M3 82% OpenCode Jun 2026 Official leaderboard
14 GLM-5.2 82% OpenCode Jun 2026 Official leaderboard
15 Claude Sonnet 5 82% Claude Code Jun 2026 Official leaderboard
16 Kimi K2.7 Code 75% OpenCode Jun 2026 Official leaderboard

The best score, 100%, costs $0.08 a task

Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

OpenAI Cost frontier One model at each effort level
0% 20% 40% 60% 80% 100% $0.01 $0.02 $0.05 $0.10 $0.20 $0.50 $1.00 Cost per task, log scale GPT-6.1 Terra, high GPT-6 Luna, high 0% 20% 40% 60% 80% 100% $0.01 $0.10 $1.00 Cost per task, log scale GPT-6.1 Terra, high GPT-6 Luna, high

Source: Official Next.js Agent Evals leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Fable 5.1 high Claude Code 100% $0.68
Claude Opus 5.5 high Claude Code 100% $0.21
GPT-6 Sol high Codex 100% $0.25
Gemini 4 Argon OpenCode 100% $0.45
GPT-6.1 Terra high Codex 100% $0.08
Grok 4.7 OpenCode 96% $0.17
Gemini 3.8 Flash OpenCode 93% $0.61
GPT-6 Astra high Codex 93% $0.84
Kimi K3 OpenCode 86% $0.27
GPT-6 Luna high Codex 86% $0.01
MiniMax-M3 OpenCode 82% $0.05
GLM-5.2 OpenCode 82% $0.15
Claude Sonnet 5 Claude Code 82% $0.35
Kimi K2.7 Code OpenCode 75% $0.13

3 models have held number one on the official leaderboard since its release in October 2025

Lab claims count from the day the card came out, independent runs from the model’s release.

MiniMax Moonshot AI Anthropic
Own leaderboard Passed 85% in Jul 2026, by Kimi K3
Claude Fable 5.1 100%

Source: Official Next.js Agent Evals leaderboard

Kimi K3 on the official leaderboard passed 85% in July 2026

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Claude Fable 5.1 Kimi K3 0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Claude Fable 5.1 Kimi K3

Source: Official Next.js Agent Evals leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Coding

All 282 in Coding
  • SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
  • LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests. DeepSeek-V4-Pro 93.5%
  • SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
  • DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
  • SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases. Seed2.1 Pro 59.8%
  • Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance. DeepSeek V4.1 Flash 3,471