qwen3.5-35b-a3b: benchmark scores, price and ranking

qwen3.5-35b-a3b from Alibaba ranks #372 of 743 LLMs on the LLMs Tiger aggregate benchmark score (48.2), based on 7 benchmark results captured up to 2026-10-13.

Organisation: Alibaba
Type: Open-weight / local model
Open weights: Yes
Released: 2026-02-24 (source: epoch)
Cheapest API price: $0.140 in / $1.00 out per million tokens (via deepinfra, observed 2026-10-13T20:47Z)
Aggregate rank: #372 of 743 (score 48.2)

Relative to other models, qwen3.5-35b-a3b is strongest on Epoch ECI (#41 of 204) and weakest on Chess Puzzles (#98 of 122).

Across the 6 benchmarks where at least ten models are scored, qwen3.5-35b-a3b is in the top 10% on 0 and in the top half on 4.

Its cheapest listed input price, $0.140 per million tokens, is lower than that of 75% of the 427 models with a listed price.

qwen3.5-35b-a3b benchmark results

qwen3.5-35b-a3b benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
Epoch ECI	143	#41 of 204	Epoch AI Benchmarking Hub	2026-10-13
GPQA Diamond	83.5%	#91 of 275	Epoch AI Benchmarking Hub	2026-10-13
DTBench	80.0%	#56 of 146	Epoch AI Benchmarking Hub	2026-10-13
LMCA	29.5%	#53 of 108	Epoch AI Benchmarking Hub	2026-10-13
OTIS Mock AIME	54.4%	#101 of 189	Epoch AI Benchmarking Hub	2026-10-13
Chess Puzzles	1.0%	#98 of 122	Epoch AI Benchmarking Hub	2026-10-13
SWE-rebench	17.1%	#7 of 7	SWE-rebench leaderboard	2026-10-13

About qwen3.5-35b-a3b

qwen3.5-35b-a3b is an open-weight model that can run on your own hardware from Alibaba. It was released on 2026-02-24, according to Epoch AI. The cheapest API listing we track charges $0.140 per million input tokens and $1.00 per million output tokens, through deepinfra. We hold 7 benchmark results for it across 4 areas, and none is in the top tenth of its board.

Where qwen3.5-35b-a3b stands

On our aggregate score, qwen3.5-35b-a3b is #372 of 743 ranked models, around the middle of the table. Its score is 48.2; the scale is centred on 50, so it sits 1.8 points below the typical model. That beats 50% of the ranked field. Among the 74 ranked Alibaba models it is #36. The next one up is Qwen3.8-27B (XHigh) at 48.5. The next one down is qwen2:72b-instruct-q8_0 at 48.2.

qwen3.5-35b-a3b, benchmark by benchmark

Epoch ECI puts qwen3.5-35b-a3b at 143, #41 of 204. That beats 80% of the models tested, but 14 points short of Qwen 3.8 Max (156). The results either side are GPT-5.5 Instant (143) and Gemini 2.5 Pro (May 2025) (142). It covers Epoch Capabilities Index: Epoch's own aggregate of ~40 benchmarks. The board median is 127, so qwen3.5-35b-a3b is 15 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

On GPQA Diamond, qwen3.5-35b-a3b reaches 83.5%. It ranks #91 of 275, ahead of 67% of the field, and 12.3 points behind GPT-6 Astra (max) at 95.8%. The results either side are Step-3.5-Flash (83.5%) and GLM-5.1-FP8 (83.3%). It covers expert-level science questions. The board median is 71.7%, so qwen3.5-35b-a3b is 11.7 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

qwen3.5-35b-a3b manages 80.0% on DTBench, good for #56 of 146. Claude Opus 5.5 (max) leads at 98.9%, 18.9 points further up, and 62% of models sit below. The results either side are GPT-5.4 mini (xhigh) (80.0%) and MiniMax-M3 (78.9%). It covers dynamic tool-use benchmark. The board median is 68.8%, so qwen3.5-35b-a3b is 11.2 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

LMCA puts qwen3.5-35b-a3b at 29.5%, #53 of 108. That beats 51% of the models tested, but 38.8 points short of Claude Opus 5.5 (max) (68.2%). The results either side are mimo-v2.5-pro (29.5%) and Qwen3-235B-A22B-Thinking-2507 (29.3%). It covers language-model capability assessment. The board median is 29.2%, so qwen3.5-35b-a3b is 0.2 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

qwen3.5-35b-a3b lands below the middle on OTIS Mock AIME: 54.4%, #101 of 189, with 53% of the field scoring higher. Claude Fable 5.1 (max) leads at 100.0%; the closest result above is grok-3-beta at 55.6%. It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so qwen3.5-35b-a3b is 3.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Chess Puzzles is a weak spot. qwen3.5-35b-a3b scores 1.0%, #98 of 122, behind 80% of the models measured. The leader, GPT-6 Astra (max), reaches 72.0%. It covers chess tactics puzzles. The board median is 12.5%, so qwen3.5-35b-a3b is 11.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

The full table above also lists the other 1 result, including boards with fewer than ten models, where a rank says little.

Strengths and weaknesses

Reasoning, maths and science: solid, ahead of 57% of the field on average across 2 benchmarks. Its best board here is GPQA Diamond (#91 of 275) and its weakest is OTIS Mock AIME (#101 of 189). General knowledge and chat: solid, ahead of 50% of the field on average across 3 benchmarks. Its best board here is Epoch ECI (#41 of 204) and its weakest is Chess Puzzles (#98 of 122).

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so qwen3.5-35b-a3b costs less than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), DeepSeek V4 Flash 0731 (max) ($0.016, 53.1). Similar scores cost more on Gemini 2.0 Flash (Experimental) ($0.150) and DeepSeek-R1-Distill-Qwen-32B ($0.150).

Release history

The Alibaba model released before it was Qwen3.5 27B on 2026-02-24, which scores 49.3. The next one was qwen3.5-4b (none) on 2026-02-24, at 49.6.

Reading these numbers

The 7 scores on this page come from 2 sources: Epoch AI Benchmarking Hub (6), SWE-rebench leaderboard (1). No official leaderboard supplied any of them, so treat them as third-party measurements. The leaderboard also lists 1 other reasoning-effort setting of this model, which redirect here. This page shows the highest. With 7 results the aggregate rank can move a lot when one new score arrives.

qwen3.5-35b-a3b is also listed at 1 other reasoning-effort setting; this page shows the highest-effort row, which is the one the leaderboard keeps by default.