Benchmark Almanac

GraphWalks

Claude Fable 5 100% leads since Jun 2026 In use

A graph is given as an edge list in long context. The model answers BFS or parent queries, scored by F1.

On 2026-02-27 OpenAI fixed wrong ground truth in 24 of 400 parents samples under 128K. It also reworded the BFS prompt to ask only for nodes at the exact depth. The name stayed, so results before and after may differ slightly.

Made by OpenAI Released Apr 2025 Reasoning over long documents All tasks public Last in a launch post Oct 2026 Last model added Oct 2026
Level
University Undergraduates and entry-level professionals
Scored
Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
Answers
Short answer A number, word or short string matched to a key.
Dataset openai.com
21
Models scored in lab cards
19
Launch posts list it, from 8 labs
13
System cards report it, from 6 labs

Claude Fable 5 leads with 100%

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

21 models. Scroll sideways for the rest.

  1. 100.0 Claude Fable 5
  2. 99.7 Gemini 4 Argon
  3. 99.3 Claude Opus 4.8
  4. 96.2 Gemini 4 Carbon
  5. 90.0 MAI-Thinking-1
  6. 89.7 Nemotron 3.5 Super VL
  7. 83.3 Seed2.0 Pro
  8. 82.3 Gemma 4 31B IT
  9. 81.3 GPT-5.6 Luna
  10. 80.0 Claude Mythos Preview
  11. 76.9 GPT-5.6 Terra
  12. 73.4 GPT-5 mini
  13. 73.4 GPT-5.4 Nano
  14. 72.6 Gemma 4 26B A4B
  15. 64.0 GPT-5 nano
  16. 63.2 Spark-X2.6-4B
  17. 61.7 GPT-4.1
  18. 61.7 GPT-4.1 mini
  19. 56.0 LongCat-Flash-Omni
  20. 51.0 LongCat-Flash-Chat
  21. 25.0 GPT-4.1 nano
  1. 1 Claude Fable 5 100%
  2. 2 Gemini 4 Argon 99.7%
  3. 3 Claude Opus 4.8 99.3%
  4. 4 Gemini 4 Carbon 96.2%
  5. 5 MAI-Thinking-1 90%
  6. 6 Nemotron 3.5 Super VL 89.7%
  7. 7 Seed2.0 Pro 83.3%
  8. 8 Gemma 4 31B IT 82.3%
  9. 9 GPT-5.6 Luna 81.3%
  10. 10 Claude Mythos Preview 80%
  11. 11 GPT-5.6 Terra 76.9%
  12. 12 GPT-5 mini 73.4%
  13. 13 GPT-5.4 Nano 73.4%
  14. 14 Gemma 4 26B A4B 72.6%
  15. 15 GPT-5 nano 64%
  16. 16 Spark-X2.6-4B 63.2%
  17. 17 GPT-4.1 61.7%
  18. 18 GPT-4.1 mini 61.7%
  19. 19 LongCat-Flash-Omni 56%
  20. 20 LongCat-Flash-Chat 51%
  21. 21 GPT-4.1 nano 25%
#ModelScoreSettingDateSource
1 Claude Fable 5 100% Jun 2026 Model card
2 Gemini 4 Argon 99.7% Oct 2026 Model card
3 Claude Opus 4.8 99.3% May 2026 Model card
4 Gemini 4 Carbon 96.2% Oct 2026 Model card
5 MAI-Thinking-1 90% Jun 2026 Model card
6 Nemotron 3.5 Super VL 89.7% Oct 2026 Model card
7 Seed2.0 Pro 83.3% Feb 2026 Model card
8 Gemma 4 31B IT 82.3% Jul 2026 Model card
9 GPT-5.6 Luna 81.3% Jul 2026 Model card
10 Claude Mythos Preview 80% Apr 2026 Model card
11 GPT-5.6 Terra 76.9% Jul 2026 Model card
12 GPT-5 mini 73.4% Aug 2025 Model card
13 GPT-5.4 Nano 73.4% Mar 2026 Model card
14 Gemma 4 26B A4B 72.6% Jul 2026 Model card
15 GPT-5 nano 64% Aug 2025 Model card
16 Spark-X2.6-4B 63.2% Oct 2026 Model card
17 GPT-4.1 61.7% Apr 2025 Model card
18 GPT-4.1 mini 61.7% Apr 2025 Model card
19 LongCat-Flash-Omni 56% Oct 2025 Model card
20 LongCat-Flash-Chat 51% Aug 2025 Model card
21 GPT-4.1 nano 25% Apr 2025 Model card

5 models have held number one since its release in Apr 2025

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

OpenAI ByteDance Seed Anthropic
Lab claims Passed 85% in May 2026, by Claude Opus 4.8
Claude Fable 5 100%

The best score passed 85% in May 2026

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

Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% 2025 2026 2027 Near the ceiling (85%) Claude Fable 5 Seed2.0 Pro GPT-5 mini GPT-4.1 0% 20% 40% 60% 80% 100% 2025 2026 2027 Near the ceiling (85%) Claude Fable 5 Seed2.0 Pro

6 labs put it in a launch post in the last six months

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

OpenAI
Meituan
ByteDance Seed
Anthropic
Microsoft
Google DeepMind
NVIDIA
iFlytek
DateLaunch postScoreAs the post writes it
28 Oct 2026 Gemini 4 Carbon 96.2 GraphWalks (256k to 1M, BFS (F1))
22 Oct 2026 Spark-X2.6-4B 63.2 Mean of Graphwalks bfs 256k f1 92.2 / Graphwalks bfs 1m f1 34.2
15 Oct 2026 Nemotron 3.5 Super VL 89.7 GraphWalks
13 Oct 2026 Gemini 4 Argon 99.7 Best of GraphWalks, Up to 128k, BFS (F1) / GraphWalks, 256k to 1M, BFS (F1)
9 Jul 2026 GPT-5.6 Luna 81.3 GraphWalks BFS 256k f1 81.3 (1mil f1 51.2)
9 Jul 2026 GPT-5.6 Terra 76.9 GraphWalks BFS 256k f1 76.9 (1mil f1 71.2)
9 Jun 2026 Claude Fable 5 100 Best of GraphWalks BFS 256K / GraphWalks Parents 256K
2 Jun 2026 MAI-Thinking-1 90 GraphWalks (≤128k)
28 May 2026 Claude Opus 4.8 99.3 Best of GraphWalks BFS 256K / GraphWalks Parents 256K
7 Apr 2026 Claude Mythos Preview 80 GraphWalks BFS 256K-1M
17 Mar 2026 GPT-5.4 Nano 73.4 Graphwalks BFS 0K-128K
14 Feb 2026 Seed2.0 Pro 83.3 Mean of Graphwalks Bfs (<128k) 68.9 / Graphwalks Parents (<128k) 97.6
31 Oct 2025 LongCat-Flash-Omni 56 GraphWalks-128k(precision)
29 Aug 2025 LongCat-Flash-Chat 51 GraphWalks-128k(precision)
7 Aug 2025 GPT-5 nano 64 Graphwalks bfs <128k 64.0 (parents <128k 43.8)
7 Aug 2025 GPT-5 mini 73.4 Graphwalks bfs <128k 73.4 (parents <128k 64.3)
14 Apr 2025 GPT-4.1 nano 25 Graphwalks bfs <128k 25.0 (parents <128k 9.4)
14 Apr 2025 GPT-4.1 mini 61.7 Graphwalks bfs <128k 61.7 (parents <128k 60.5)
14 Apr 2025 GPT-4.1 61.7 Best of Graphwalks bfs <128k / Graphwalks bfs >128k / Graphwalks parents <128k / Graphwalks parents >128k

13 system cards report it, from 6 labs

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
15 Oct 2026 Nemotron 3.5 Super VL technical report 89.7 GraphWalks (<=1M), BFS and parents, 950 prompts, F1 (Table 10) Capabilities
13 Oct 2026 Gemini 4 Argon model card 99.7 GraphWalks, up to 128k, BFS (F1) (results table, p.5). 256k to 1M BFS (F1) = 84.2% Capabilities
24 Jul 2026 Gemma 4 31B IT technical report 82.3 GraphWalks F1, <128k, without thinking (Table 9) Capabilities
24 Jul 2026 Gemma 4 26B A4B technical report 72.6 GraphWalks F1, <128k, without thinking (Table 9) Capabilities
2 Jun 2026 MAI-Thinking-1 technical report 90 GraphWalks (<=128k), F1 (Table 12, rounded) Capabilities
28 May 2026 Claude Opus 4.8 system card GraphWalks F1, five trials, no overall: BFS 256K 85.9, Parents 256K 99.3, BFS 1M 68.1, Parents 1M 83.3 (Table 8.1.A, Table 8.9.A) Capabilities
16 Apr 2026 Claude Opus 4.7 system card GraphWalks F1, no overall: BFS 256K-1M 58.6, parents 256K-1M 75.1, 256K subset BFS 76.91 and parents 93.57 (8.7.1) Capabilities
7 Apr 2026 Claude Mythos Preview system card GraphWalks F1, BFS 256K-1M 80.0 and parents 256K-1M 97.7, average of 5 trials, no overall (Table 6.3.A, 6.9) Capabilities
17 Feb 2026 Claude Sonnet 4.6 system card GraphWalks F1 at max effort, BFS 1M 73.8, BFS 256K subset 74.5, Parents 1M 86.4 (row label garbled in table), Parents 256K subset 97.9, no overall (Table 2.16.B) Capabilities
14 Feb 2026 Seed2.0 Pro model card Graphwalks Bfs (<128k) 68.9, Graphwalks Parents (<128k) 97.6, no overall, in-house tokenization (Table 3) Capabilities
5 Feb 2026 Claude Opus 4.6 system card GraphWalks F1 at max effort, BFS 1M 38.7, BFS 256K subset 61.1, Parents 1M 72.0, Parents 256K subset 95.4, no overall (Table 2.18.A) Capabilities
31 Oct 2025 LongCat-Flash-Omni technical report 56 GraphWalks-128K (precision) (Table 13) Capabilities
1 Sep 2025 LongCat-Flash-Chat technical report 51 GraphWalks-128k (precision) (Table 3) Capabilities

More in Long context

All 66 in Long context
  • MRCR v2 In use Multi-round coreference: retrieve the n-th of several similar requests hidden in a long synthetic conversation, up to 1M tokens. Muse Spark 1.3 98.5%
  • LongBench v2 In use 503 multiple-choice questions over 8K-2M-word contexts (QA, dialogue, code repos, structured data). Accuracy. Beam 1.0 66.9%
  • OpenAI MRCR v2 In use Models read a long synthetic chat that hides 2, 4 or 8 identical writing requests and must reproduce the response to the i-th one. Scored by mean string match ratio per context bin. K-EXAONE 2.0 94.4%
  • CL-bench In use Model must learn new knowledge from a supplied long context (rules, procedures, data) and solve tasks. Strict LLM-verified rubrics. GPT-5.4 27.9%
  • AA-LCR 1.1 In use Models answer 100 questions that need reasoning across sets of real documents of about 100K tokens. An LLM equality checker with a fixed system prompt compares each answer with the key. Step 5 Preview 88.3%
  • LOCA-Bench In use Agentic tasks whose environment state is scaled to grow context length while task semantics stay fixed. Success rate versus context length. GPT-5.2 21.3%