Claude Fable 5.1 up to #10, Claude Fable 5 up to #9 (28 Oct)

BenchLeader refreshed 1632 models from 24 of 24 data feeds (benchmark publishers, aggregators and price lists). 11 top-10 rank changes, 171 price changes, 1292 new or updated benchmark results across 24 data feeds.

Movement in the top 10

GPT-6 Astra moved from #9 to #3 in the BenchLeader Index. (details)
GPT-6 Astra (high) moved from #6 to #5 in the BenchLeader Index. (details)
GPT-6 Astra (xhigh) moved from #14 to #6 in the BenchLeader Index. (details)
Claude Fable 5.1 (max) moved from #3 to #7 in the BenchLeader Index. (details)
Claude Opus 5.5 (xhigh) moved from #12 to #8 in the BenchLeader Index. (details)
Claude Fable 5 moved from #20 to #9 in the BenchLeader Index. (details)
Claude Fable 5.1 moved from #22 to #10 in the BenchLeader Index. (details)
Gemini 4 Argon (high) moved from #5 to #11 in the BenchLeader Index. (details)
Claude Opus 5.5 (high) moved from #8 to #12 in the BenchLeader Index. (details)
Claude Fable 5 (max) moved from #7 to #16 in the BenchLeader Index. (details)
Claude Fable 5.1 (xhigh) moved from #10 to #20 in the BenchLeader Index. (details)

Price changes

Claude Opus 5.5 (max) cost per task moved from $5.98/M to $10.04/M (68% up). (details)
GPT-6 Astra (max) cost per task moved from $3.26/M to $6.24/M (92% up). (details)
GPT-6.1 Sol (max) cost per task moved from $0.72/M to $1.27/M (76% up). (details)
GPT-6 Astra (high) cost per task moved from $1.73/M to $3.24/M (88% up). (details)
GPT-6 Astra (xhigh) cost per task moved from $2.31/M to $3.66/M (58% up). (details)
Claude Fable 5.1 (max) cost per task moved from $7.63/M to $13.04/M (71% up). (details)
Claude Opus 5.5 (xhigh) cost per task moved from $3.46/M to $6.05/M (75% up). (details)
Gemini 4 Argon (high) cost per task moved from $1.62/M to $3.84/M (137% up). (details)
Claude Opus 5.5 (high) cost per task moved from $1.82/M to $3.46/M (90% up). (details)
Claude Sonnet 5.5 (max) cost per task moved from $5.46/M to $10.02/M (83% up). (details)
GPT-6 Astra (medium) cost per task moved from $1.54/M to $2.73/M (77% up). (details)
GPT-6.1 Terra (max) cost per task moved from $0.40/M to $0.96/M (138% up). (details)
Claude Fable 5.1 (xhigh) cost per task moved from $5.98/M to $10.89/M (82% up). (details)
GPT-6.1 Sol (xhigh) cost per task moved from $0.39/M to $0.76/M (94% up). (details)
GPT-6.1 Sol (high) cost per task moved from $0.32/M to $0.59/M (85% up). (details)
Claude Fable 5.1 (high) cost per task moved from $3.91/M to $7.56/M (93% up). (details)
Claude Sonnet 5.5 (xhigh) cost per task moved from $2.01/M to $4.09/M (103% up). (details)
GPT-6.1 Sol (medium) cost per task moved from $0.21/M to $0.36/M (68% up). (details)
Claude Opus 5.5 (medium) cost per task moved from $1.34/M to $2.45/M (83% up). (details)
GPT-6 Astra (low) cost per task moved from $0.82/M to $1.42/M (74% up). (details)
Claude Fable 5.1 (medium) cost per task moved from $2.98/M to $5.58/M (87% up). (details)
Muse Wonder (max) cost per task moved from $3.27/M to $6.95/M (113% up). (details)
Muse Spark 1.3 (max) cost per task moved from $1.31/M to $2.39/M (81% up). (details)
GPT-6.1 Terra (xhigh) cost per task moved from $0.19/M to $0.40/M (107% up). (details)
Gemini 3.8 Flash (high) cost per task moved from $3.49/M to $7.77/M (123% up). (details)
GLM 5.2 (max) cached input price moved from $0.30/M to $0.26/M (13% down). (details)
GLM-5 (thinking) cached input price moved from $0.15/M to $0.20/M (34% up). (details)
GPT-6.1 Terra (high) cost per task moved from $0.13/M to $0.26/M (93% up). (details)
Kimi K3.1 Pro cost per task moved from $2.11/M to $3.68/M (74% up). (details)
Qwen3.8 Max (max) cost per task moved from $5.41/M to $10.02/M (85% up). (details)
…and 141 more.

New benchmark results

AA Progress Index v1: 630 new results, including Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol… (board)
HLE-Diamond (AA): 132 new results, including Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol… (board)
CritPt v2 hard subset (AA): 86 new results, including Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol… (board)
FrontierCode 2.0 (AA): 77 new results, including Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol… (board)
SAGE: 7 new results, including DeepSeek V4.1 Flash, GLM 5.3 Flash, MiniMax M3.1 Flash… (board)
MLCR: 2 new results, including GPT-6.1 Terra, Muse Wonder Full (board)
Terminal-Bench-Science 0.1 (AA): 2 new results, including Ling 3.1 Flash, Mistral Small 4 (board)
Arena WebDev: 1 new result, including GPT-6.1 Terra (board)
PRBench Finance: 1 new result, including DeepSeek V4.1 Flash 1017 (board)
PRBench Legal: 1 new result, including DeepSeek V4.1 Flash 1017 (board)

Score revisions

Claude Opus 5.5 (max) on AA Intelligence Index v5.0: 57.6 → 58.4. (details)
Claude Opus 5.5 (max) on AA-Omniscience v1.1: 46.4 → 48.1. (details)
Claude Opus 5.5 (max) on AA-Omniscience v1.1: accuracy: 66.2% → 65.5%. (details)
Claude Opus 5.5 (max) on AA-Omniscience v1.1: non-hallucination: 41.4% → 49.7%. (details)
GPT-6 Astra (max) on AA Intelligence Index v5.0: 52.7 → 56.9. (details)
GPT-6 Astra (max) on AA-Omniscience v1.1: 43.4 → 42.4. (details)
GPT-6 Astra (max) on AA-Omniscience v1.1: accuracy: 62.6% → 61.6%. (details)
GPT-6 Astra (max) on AA-Omniscience v1.1: non-hallucination: 48.7% → 50.2%. (details)
GPT-6.1 Sol (max) on AA Intelligence Index v5.0: 51.8 → 54.5. (details)
GPT-6.1 Sol (max) on AA-Omniscience v1.1: 41.5 → 42. (details)
GPT-6.1 Sol (max) on AA-Omniscience v1.1: non-hallucination: 45.7% → 48.1%. (details)
GPT-6.1 Sol (max) on SAGE: 46.5% → 90.5%. (details)
GPT-6 Astra (high) on AA Intelligence Index v5.0: 50.9 → 54.9. (details)
GPT-6 Astra (high) on AA-Omniscience v1.1: 43.7 → 44.6. (details)
GPT-6 Astra (high) on AA-Omniscience v1.1: non-hallucination: 55.2% → 57.5%. (details)
GPT-6 Astra (xhigh) on AA Intelligence Index v5.0: 52.4 → 55.3. (details)
GPT-6 Astra (xhigh) on AA-Omniscience v1.1: accuracy: 61.9% → 61%. (details)
GPT-6 Astra (xhigh) on AA-Omniscience v1.1: non-hallucination: 51.7% → 55.4%. (details)
Claude Fable 5.1 (max) on AA Intelligence Index v5.0: 53.4 → 49. (details)
Claude Fable 5.1 (max) on AA-Omniscience v1.1: 43.5 → 46.3. (details)
…and 333 more.

Speed changes

GPT-6 Astra time to first token changed from 1.98 s to 4.91 s. (details)
Claude Fable 5.1 time to first answer changed from 207.46 s to 132.23 s. (details)
Claude Fable 5.1 response time changed from 215.04 s to 139.76 s. (details)
GPT-6.1 Terra time to first answer changed from 189.33 s to 77.96 s. (details)
GPT-6.1 Terra response time changed from 194.83 s to 83.46 s. (details)
Claude Opus 4.8 time to first token changed from 0.74 s to 1.61 s. (details)
GPT-5.6 Terra time to first answer changed from 16.82 s to 9.98 s. (details)
GPT-5.2 time to first token changed from 1.24 s to 2.17 s. (details)
GLM 5.2 time to first token changed from 0.92 s to 0.52 s. (details)
Grok 4.20 Multi-Agent time to first token changed from 4.44 s to 7.65 s. (details)
GLM 5.3 time to first token changed from 0.39 s to 0.56 s. (details)
DeepSeek V4 Flash time to first token changed from 0.95 s to 1.30 s. (details)
GPT-5.2-Codex output speed changed from 17 tok/s to 46 tok/s. (details)
GPT-5.2-Codex time to first token changed from 3.90 s to 1.95 s. (details)
GLM 5.1 output speed changed from 29 tok/s to 41 tok/s. (details)
…and 45 more.

Today's top five

More: New AI model releases this week, and the full LLM leaderboard 2026.

This digest is generated automatically from the day's data changes. Every line links to the page where you can check the numbers and their source.