MiniMax M3.1 Flash added, Mercury 3 added (19 Oct 2026)

BenchLeader refreshed 1601 models from 24 of 24 data feeds (benchmark publishers, aggregators and price lists). 2 new models, 10 top-10 rank changes, 77 price changes, 384 new or updated benchmark results across 24 data feeds.

New models

MiniMax-M3.1-Flash (MiniMax) appeared with 5 benchmark results, entering the BenchLeader Index at #85. (details)
Mercury 3 (Inception) is now listed at $0.16/M blended; no independent results yet. (details)

Movement in the top 10

GPT-6 Astra (max) moved from #4 to #3 in the BenchLeader Index. (details)
Claude Fable 5.1 (max) moved from #3 to #4 in the BenchLeader Index. (details)
GPT-6.1 Sol (max) moved from #10 to #5 in the BenchLeader Index. (details)
Claude Opus 5.5 (xhigh) moved from #9 to #6 in the BenchLeader Index. (details)
GPT-6 Astra (high) moved from #5 to #7 in the BenchLeader Index. (details)
Claude Sonnet 5.5 (max) moved from #21 to #8 in the BenchLeader Index. (details)
Claude Fable 5.1 (xhigh) moved from #6 to #9 in the BenchLeader Index. (details)
Claude Fable 5 (max) moved from #12 to #10 in the BenchLeader Index. (details)
Claude Opus 5.5 (high) moved from #7 to #11 in the BenchLeader Index. (details)
Claude Fable 5.1 (high) moved from #8 to #12 in the BenchLeader Index. (details)

Price changes

Qwen3.8 Max (max) cached input price moved from $0.25/M to $0.22/M (11% down). (details)
Qwen3.7 Plus cached input price moved from $0.040/M to $0.050/M (25% up). (details)
Qwen3.5 Plus 2026-04-20 (thinking) input price moved from $0.40/M to $0.12/M (71% down). (details)
Qwen3.5 Plus 2026-04-20 (thinking) output price moved from $2.40/M to $0.69/M (71% down). (details)
Qwen3.5 Flash input price moved from $0.10/M to $0.029/M (71% down). (details)
Qwen3.5 Flash output price moved from $0.40/M to $0.29/M (28% down). (details)
Qwen3 32B input price moved from $0.70/M to $0.29/M (59% down). (details)
Qwen3 32B output price moved from $2.80/M to $1.15/M (59% down). (details)
Qwen3 235B A22B input price moved from $0.70/M to $0.29/M (59% down). (details)
Qwen3 235B A22B output price moved from $2.80/M to $1.15/M (59% down). (details)
Qwen-Plus input price moved from $0.40/M to $0.12/M (71% down). (details)
Qwen-Plus output price moved from $1.20/M to $0.29/M (76% down). (details)
Qwen3 14B input price moved from $0.35/M to $0.14/M (59% down). (details)
Qwen3 14B output price moved from $1.40/M to $0.57/M (59% down). (details)
Qwen3 235B A22B (thinking) input price moved from $0.70/M to $0.29/M (59% down). (details)
Qwen3 235B A22B (thinking) output price moved from $2.80/M to $1.15/M (59% down). (details)
Qwen3 235B A22B (no reasoning) input price moved from $0.70/M to $0.29/M (59% down). (details)
Qwen3 235B A22B (no reasoning) output price moved from $2.80/M to $1.15/M (59% down). (details)
Qwen2.5 7B Instruct input price moved from $0.17/M to $0.072/M (59% down). (details)
Qwen2.5 7B Instruct output price moved from $0.70/M to $0.14/M (79% down). (details)
Qwen3 8B (thinking) input price moved from $0.18/M to $0.072/M (60% down). (details)
Qwen3 8B (thinking) output price moved from $0.70/M to $0.29/M (59% down). (details)
Qwen2.5 32B Instruct input price moved from $0.70/M to $0.29/M (59% down). (details)
Qwen2.5 32B Instruct output price moved from $2.80/M to $0.86/M (69% down). (details)
Qwen3 8B input price moved from $0.18/M to $0.072/M (60% down). (details)
Qwen3 8B output price moved from $0.70/M to $0.29/M (59% down). (details)
Qwen2.5 VL 72B Instruct input price moved from $2.80/M to $2.29/M (18% down). (details)
Qwen2.5 VL 72B Instruct output price moved from $8.40/M to $6.88/M (18% down). (details)
Qwen3.5 Plus 2026-04-20 input price moved from $0.40/M to $0.12/M (71% down). (details)
Qwen3.5 Plus 2026-04-20 output price moved from $2.40/M to $0.69/M (71% down). (details)
…and 47 more.

New benchmark results

LiveBench 2.0: 42 new results, including GPT-6 Astra, Claude Fable 5, Claude Fable 5.1… (board)
LiveBench Long Agentic: 42 new results, including GPT-6 Astra, Claude Fable 5, Claude Fable 5.1… (board)
LiveBench 2.0 Coding: 42 new results, including GPT-6 Astra, Claude Fable 5, Claude Fable 5.1… (board)
LiveBench Browser: 42 new results, including GPT-6 Astra, Claude Fable 5, Claude Fable 5.1… (board)
LiveBench Research: 42 new results, including GPT-6 Astra, Claude Fable 5, Claude Fable 5.1… (board)
LiveBench Design: 42 new results, including GPT-6 Astra, Claude Fable 5, Claude Fable 5.1… (board)
LiveBench 2.0 Reasoning: 34 new results, including GPT-6 Astra, Claude Fable 5, Claude Fable 5.1… (board)
LiveBench 2.0 Data Analysis: 34 new results, including GPT-6 Astra, Claude Fable 5, Claude Fable 5.1… (board)
LMCA: 20 new results, including Gemini 4 Argon, GPT-6.1 Sol, Claude Opus 5.5… (board)
DTBench: 20 new results, including Gemini 4 Argon, GPT-6.1 Sol, Claude Opus 5.5… (board)
ALE-Bench: 3 new results, including Gemini 4 Argon, Muse Spark Open 816B A42B, Gemini 4 Argon (board)
FORTRESS: 1 new result, including Claude Opus 5.5 (board)

Score revisions

Kimi K3 (max) on ForecastBench: 61.9% → 61.1%. (details)
MiMo-V2.6-Pro on APEX-Agents: 59.5% → 54.4%. (details)
Claude Opus 4.7 on ForecastBench: 59.9% → 59.4%. (details)
DeepSeek V4.1 Flash (max) on Arena WebDev: 1618 → 1670. (details)
Claude Opus 4.8 on ForecastBench: 59.8% → 59.1%. (details)
Kimi K3 on APEX-Agents: 50.6% → 47.8%. (details)
Muse Spark 1.3 on APEX-Agents: 57.8% → 58.6%. (details)
Claude Sonnet 4.6 on ForecastBench: 61.1% → 60.6%. (details)
Claude Sonnet 5 on ForecastBench: 60.4% → 59.4%. (details)
Gemini 3.5 Flash Lite on Arena WebDev: 1441 → 1435. (details)
Llama 4 Maverick on SciCode: 33.1% → 31.7%. (details)
Grok 4.7 on ProofBench: 34% → 26%. (details)
Muse Spark 1.2 on APEX-Agents: 36.4% → 45.4%. (details)
Muse Spark Open (max) on Arena WebDev: 1511 → 1524. (details)

Speed changes

GPT-6 Astra time to first answer changed from 88.82 s to 49.58 s. (details)
GPT-6 Astra response time changed from 98.10 s to 58.13 s. (details)
Claude Opus 5 time to first answer changed from 78.72 s to 44.14 s. (details)
Claude Opus 5 response time changed from 87.33 s to 53.45 s. (details)
Claude Opus 5 time to first token changed from 5.70 s to 3.21 s. (details)
Claude Fable 5.1 time to first answer changed from 14.55 s to 3.19 s. (details)
Claude Fable 5.1 response time changed from 22.96 s to 12.15 s. (details)
GPT-6 Sol time to first token changed from 1.42 s to 2.66 s. (details)
GPT-5.6 Sol time to first answer changed from 22.60 s to 9.97 s. (details)
GPT-5.6 Sol response time changed from 28.33 s to 15.84 s. (details)
GPT-5.5 time to first answer changed from 31.36 s to 17.02 s. (details)
GPT-5.5 response time changed from 36.46 s to 22.59 s. (details)
Claude Opus 5.5 time to first answer changed from 10.38 s to 4.90 s. (details)
GPT-5.6 Terra time to first token changed from 1.29 s to 2.63 s. (details)
GPT-6 Sol time to first answer changed from 54.14 s to 32.66 s. (details)
…and 89 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.