State of the leaderboard: week 43, 19–25 October 2026

Complete week: data as of 25 Oct 2026, from 7 daily updates

The week in 6 sentences

In week 43 of 2026 (19–25 October 2026), Claude Opus 5.5 (Anthropic) held the #1 spot on the BenchLeader Index at 69.2, 1.0 points ahead of Gemini 4 Argon (68.2).
1 model entered the top 10 (GPT-5.5 at #10), pushing out Muse Spark; the rest kept their places.
10 new models arrived; the highest-ranked is Kimi K3.1 Pro (Moonshot AI) at #28 with an index of 62.4.
The biggest riser was Gemini 1.5 Pro, up 4.5 points to 47.2; the biggest faller DeepSeek V3.1, down 3.7 to 49.1.
55 models cut a list price and 61 models raised one; the largest cut was Qwen-MT Plus's output price, 89% lower.
1,149 new benchmark results were published, the most on MLCR (98).

Top 10 at the end of the week

The BenchLeader Index, one row per model at its best reasoning effort, against the table at the end of the previous week. Left the top 10: Muse Spark.

#	Model	Provider	Index	Last week	Move
1	Claude Opus 5.5	Anthropic	69.2	1	–
2	Gemini 4 Argon	Google	68.2	2	–
3	GPT-6 Astra	OpenAI	67.9	3	–
4	Claude Fable 5.1	Anthropic	67.8	4	–
5	GPT-6.1 Sol	OpenAI	67.4	5	–
6	Claude Fable 5	Anthropic	67.0	6	–
7	Claude Opus 5	Anthropic	66.4	7	–
8	GPT-5.6 Sol	OpenAI	65.6	8	–
9	Claude Sonnet 5.5	Anthropic	65.0	9	–
10	GPT-5.5	OpenAI	64.7	–	new

New models and their first results

MiniMax M3.1 Flash (MiniMax): #79 on the index at 58.0. First results: 38.6% on FrontierCode 1.1 (AA); 1588 on Arena WebDev; 73.1% on LiveBench 2.0; 34.7% on LiveBench Long Agentic.
Mercury 3 (Inception): not enough independent results for an index yet
gpt-5-6-sol-keenable@max: not enough independent results for an index yet
Kimi K3.1 Pro (Moonshot AI): #28 on the index at 62.4. First results: 47.1% on FrontierCode 1.1 (AA); 64.1 on AA Cyber Index; 60.8% on CWE-Bench (AA); 36.7% on DeepSec-Bench (AA).
Nemotron 3.5 Bolt E4B 4GB (NVIDIA): not enough independent results for an index yet
Nemotron Super VL 3.5 NVFP4 (NVIDIA): not enough independent results for an index yet
Smaug Nemotron 3.5 NVFP4 (NVIDIA): not enough independent results for an index yet
Spark X2.6 (Synlan): not enough independent results for an index yet
Kimi K3.1 Pro Highspeed (Moonshot AI): not enough independent results for an index yet
Qwen3.8 Max Internal (Alibaba): not enough independent results for an index yet

Biggest movers on the index

Model	Provider	19 Oct 2026	25 Oct 2026	Change
Gemini 1.5 Pro	Google	42.7	47.2	+4.5
DeepSeek V3.1	DeepSeek	52.8	49.1	-3.7
Nemotron 3 Ultra 550B A55B	NVIDIA	52.2	55.4	+3.2
Nemotron 3.5 Super VL 120B A12B	NVIDIA	53.6	50.4	-3.2
Beam 1.0	Reflection	50.9	47.7	-3.2
Gemini 1.5 Pro 002	Google	44.5	41.3	-3.2
Magistral Medium	Mistral AI	38.9	41.8	+2.9
Gemini 1.5 Flash 002	Google	40.4	38.2	-2.2
DeepSeek V4 Flash	DeepSeek	59.3	57.2	-2.1
Command R	Cohere	36.7	38.6	+1.9

Price changes

Cheaper (55)

Qwen-MT Plus output price moved from $7.37/M to $0.78/M (89% down).
Qwen2.5 VL 72B Instruct output price moved from $8.40/M to $1.00/M (88% down).
Qwen3-Omni Flash input price moved from $0.43/M to $0.058/M (87% down).
Qwen Vl (max) output price moved from $3.20/M to $0.57/M (82% down).
Quasar 438B (max) input price moved from $0.60/M to $0.11/M (82% down).
Mistral Nemo input price moved from $0.15/M to $0.029/M (81% down).
Qwen2.5 7B Instruct output price moved from $0.70/M to $0.14/M (79% down).
Qwen-Plus output price moved from $1.20/M to $0.29/M (76% down).
QwQ Plus output price moved from $2.40/M to $0.57/M (76% down).
Qwen3.5 Plus 2026-04-20 (thinking) output price moved from $2.40/M to $0.69/M (71% down).
Qwen3.5 Flash input price moved from $0.10/M to $0.029/M (71% down).
Qwen2.5 32B Instruct output price moved from $2.80/M to $0.86/M (69% down).
…and 43 more.

More expensive (61)

Qwen-MT Plus output price moved from $0.78/M to $7.37/M (851% up).
Qwen-VL (max) output price moved from $0.57/M to $3.20/M (457% up).
Qwen2.5 7B Instruct output price moved from $0.14/M to $0.70/M (386% up).
Qwen3-Omni Flash input price moved from $0.058/M to $0.26/M (345% up).
Qwen-Plus output price moved from $0.29/M to $1.20/M (318% up).
QwQ Plus output price moved from $0.57/M to $2.40/M (318% up).
Qwen3.5 Plus (thinking) output price moved from $0.69/M to $2.40/M (249% up).
Qwen3.5 Flash input price moved from $0.029/M to $0.10/M (245% up).
Qwen2.5 32B Instruct output price moved from $0.86/M to $2.80/M (225% up).
Qwen2.5 14B Instruct output price moved from $0.43/M to $1.40/M (225% up).
DeepSeek V4 Flash (no reasoning) output price moved from $0.19/M to $0.60/M (216% up).
Nemotron 3 Super 120B A12B output price moved from $0.90/M to $2.40/M (167% up).
…and 49 more.

List prices per million tokens; each model's full price history is on its page and the LLM API pricing tracker.

Other weeks

← Week 44, 26 Oct–1 Nov 2026 · Week 42, 12–18 Oct 2026 → · The weekly review on the blog

Benchmark categories: AI agent benchmarks · AI coding benchmarks · Reasoning and math benchmarks · Multimodal AI benchmarks · New AI model releases

How this is built: the BenchLeader Index is recomputed with every update from independent benchmark publishers (see the methodology), and a maker's own launch numbers are never used. Past ranks are rebuilt from the daily index history of the models listed today.

Cite as: BenchLeader, “AI leaderboard week 43, 19–25 Oct 2026: who was #1”, data as of 25 Oct 2026, https://www.benchleader.com/state/2026-w43.