Sponsored byFusecodeEnterprise coding API for Claude Code, Codex, and model workflows.

Free
Models
Providers
Docs

Home
Models
Qwen3.5.35b A3b

LMSpeed

The best API speed test tool

GitHub Twitter Email

Product

Features
Pricing
FAQ

Leaderboard

Overview
Speed Ranking
Latency Ranking
Health Ranking
Model Pricing
Model Speed
Reasoning
Coding

Models

All Models
GPT
Claude
Gemini
DeepSeek
Llama
Qwen

Free Models

All Free Models
Free GPT
Free Claude
Free Gemini
Free DeepSeek
Free Llama
Free Qwen

Tools

Speed Test
Provider Audit

Resources

Provider Directory
Documentation
GEO Checker
Viora
Botab
VidBee

Legal

Cookie Policy
Privacy Policy
Terms of Service

© 2026 LMSpeed All Rights Reserved.Made by Nexmoe with ❤️

Qwen3.5-35B-A3B API Pricing & Benchmarks | LMSpeed

Qwen

Qwen3.5-35B-A3B API Benchmarks, Pricing & Provider Data

The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inf...

Quality

#53of 96

54.0

LMSpeed score

Category Performance

Scores across benchmark categories on a 0-100 scale.

Coverage: 7 / 8
Avg. score: 63.2

#1Instruction following

82.2

#2Multilingual

81

#3Multimodal

80.6

#4Agents

54.8

#5

Specifications

Input and output token limits for this model, plus how it ranks on long-context understanding.

INPUT

262.1Ktokens

≈ 314.6 pages of text

OUTPUT

81.9Ktokens

8K128K1M4M

262.1K

Features

Technical Details

Input

Rankings

Excels at

It's decent at

Falls behind in

Detailed scores

Updated: Jul 9, 2026

Overall

1 metric

Overall score54.0#53 / 96

Overall score

LMSpeed rank: #53 / 96Updated: Jul 9, 2026Confidence: 3

5055.0

5055.0

5055.0

5354.0

53Qwen3.5-35B-A3B54.0

5553.0

5649.0

5649.0

5848.0

5848.0

Speed & latency

2 metrics

Output speed176.8 tok/s#20 / 113Time to first token1.27 s#60 / 113

Output speed

Source: Artificial Analysis / Qwen3.5 35B A3B (Non-reasoning)LMSpeed rank: #20 / 113Updated: Jul 9, 2026Confidence: 4

16185.5 tok/s

Pricing

2 metrics

Input price$0.250/M#33 / 138Output price$2.00/M#51 / 138

Input price

Source: Artificial Analysis / Qwen3.5 35B A3B (Non-reasoning)LMSpeed rank: #33 / 138Updated: Jul 9, 2026Confidence: 4

33$0.250/M

Reasoning & knowledge

17 metrics

MMLU-Pro85.3%#22 / 113GPQA81.9%#52 / 152

MMLU-Pro

LMSpeed rank: #22 / 113Updated: Jul 9, 2026Confidence: 3

1886.0%

18

Coding & agents

12 metrics

SciCode29.3%#113 / 149Coding score61.6#46 / 83

SciCode

Source: Artificial Analysis / Qwen3.5 35B A3B (Non-reasoning)LMSpeed rank: #113 / 149Updated: Jul 9, 2026Confidence: 4

10930.6%

Language & media

8 metrics

Multilingual score67.2#22 / 59MMLU-ProX81.0#8 / 9

Multilingual score

LMSpeed rank: #22 / 59Updated: Jul 9, 2026Confidence: 3

1970.6

21

Related model comparisons

GGPT-5.4 vs Qwen3.5-35B-A3B

3 verified comparison points

GGPT-5 vs Qwen3.5-35B-A3B

3 verified comparison points

CClaude Opus 4.6 vs Qwen3.5-35B-A3B

3 verified comparison points

Data as of Jul 9, 2026, 08:19 AMRankings are based on community-submitted tests and periodic health probes. Advisory only, not official data.Standard benchmark data may include BenchLM and other public sources.

Build with LMSpeed Data

Free

Free public API for LLM pricing, benchmarks & provider data

Real-time pricing & availability

Speed & latency benchmarks

300+ models, 600+ providers

1,000 requests/day

View API Documentation

Reasoning

50.4

#6Knowledge

50.2

#7Coding

43.3

No data:Math

Output

Total parameters

35B

Active parameters

3B

Documentation

Tokenizer

Qwen3

Architecture

text+image+video->text

Moderated

No

Supported parameters

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p

17GPT-5.1 Codex183.7 tok/s

18Step 3.5 Flash179.5 tok/s

19Nemotron 3 Ultra179.1 tok/s

20Qwen3.5-35B-A3B176.8 tok/s

21GPT-5 Codex169.4 tok/s

22Qwen3.5-122B-A10B163.6 tok/s

23Grok 4.20162.7 tok/s

24Kimi K2 Thinking161.7 tok/s

25GPT-5.4 Mini161.3 tok/s

Time to first token

Source: Artificial Analysis / Qwen3.5 35B A3B (Non-reasoning)LMSpeed rank: #60 / 113Updated: Jul 9, 2026Confidence: 4

56Qwen3 VL 235B A22B Instruct1.24 s

57MiniMax M21.25 s

58Qwen3.5 Omni Plus1.26 s

59Qwen3 235B A22B1.26 s

60Qwen3.5-35B-A3B1.27 s

611.28 s

621.32 s

631.33 s

641.35 s

651.35 s

33Gemini 3.1 Flash Lite Preview$0.250/M

33GPT-5 Mini$0.250/M

33GPT-5.1 Codex Mini$0.250/M

33Qwen3.5-35B-A3B$0.250/M

38DeepSeek V3.1 Terminus$0.270/M

39DeepSeek V3.2$0.280/M

39DeepSeek V3.2 Exp$0.280/M

41Gemini 2.5 Flash$0.300/M

41GLM-4.6V$0.300/M

Output price

Source: Artificial Analysis / Qwen3.5 35B A3B (Non-reasoning)LMSpeed rank: #51 / 138Updated: Jul 9, 2026Confidence: 4

51GPT-5.1 Codex Mini$2.00/M

51MiMo-V2-Omni$2.00/M

51Mistral Medium 3.1$2.00/M

51Qwen3 Next 80B A3B Instruct$2.00/M

51Qwen3.5-35B-A3B$2.00/M

57$2.10/M

58$2.20/M

58$2.20/M

58$2.20/M

58$2.20/M

86.0%

21GLM-585.7%

22O385.3%

22Qwen3.5-35B-A3B85.3%

24Qwen3.6 35B A3B85.2%

25DeepSeek R184.9%

26Kimi K2 Thinking84.8%

27Qwen3 235B A22B Instruct 250784.3%

28O184.1%

GPQA

Source: Artificial Analysis / Qwen3.5 35B A3B (Non-reasoning)LMSpeed rank: #52 / 152Updated: Jul 9, 2026Confidence: 4

48GPT-5.4 Mini82.3%

49Gemini 2.5 Pro Preview 06-0582.2%

49Gemini 3.1 Flash Lite Preview82.2%

51GLM-582.0%

52Qwen3.5-35B-A3B81.9%

5381.7%

5481.3%

5481.3%

5681.2%

5781.0%

HLE

Source: Artificial Analysis / Qwen3.5 35B A3B (Non-reasoning)LMSpeed rank: #61 / 150Updated: Jul 9, 2026Confidence: 4

57Qwen3.5 Omni Plus13.9%

58GLM-4.613.3%

59Qwen3.5-27B13.2%

60Claude Opus 4.512.9%

61Qwen3.5-35B-A3B12.8%

6212.5%

6312.3%

6312.3%

6512.2%

6611.7%

Reasoning score

LMSpeed rank: #50 / 65Updated: Jul 9, 2026Confidence: 3

46Llama 4 Scout39.7

47Qwen3.5-122B-A10B39.6

48Qwen3.5-27B39.2

49GPT-4 Turbo37.0

50Qwen3.5-35B-A3B36.2

5135.8

5233.8

5329.2

5329.2

5528.5

LongBench v2

LMSpeed rank: #8 / 8Updated: Jul 9, 2026Confidence: 3

1Claude Opus 4.564.4

3Qwen3.6 Plus62.0

5GLM-560.8

660.6

760.2

8Qwen3.5-35B-A3B59.0

AA-LCR

LMSpeed rank: #37 / 67Updated: Jul 9, 2026Confidence: 3

34Qwen3.6 35B A3B63.7

35GLM-563.3

36DeepSeek V4 Flash63.0

37Qwen3.5-35B-A3B62.7

3862.3

3961.0

3961.0

4160.7

4160.7

CritPt

LMSpeed rank: #43 / 54Updated: Jul 9, 2026Confidence: 3

37Qwen3.6 27B1.1

43Claude Sonnet 4.60.9

43DeepSeek V3.20.9

43Qwen3.5-27B0.9

43Qwen3.5-35B-A3B0.9

470.6

470.6

470.6

500.3

500.3

Knowledge score

LMSpeed rank: #26 / 79Updated: Jul 9, 2026Confidence: 3

22GPT-5.4 Mini71.3

23MiMo-V2-Pro70.9

24Qwen3.5-27B70.8

25Claude Sonnet 4.569.9

26Qwen3.5-35B-A3B69.1

2768.0

2867.7

2867.7

3067.0

3165.5

SuperGPQA

LMSpeed rank: #13 / 14Updated: Jul 9, 2026Confidence: 3

5Claude Opus 4.570.6

8Qwen3.5-122B-A10B67.1

9GLM-566.8

1066.0

1165.6

1264.7

13Qwen3.5-35B-A3B63.4

1462.6

Artificial Analysis Intelligence Index

LMSpeed rank: #43 / 77Updated: Jul 9, 2026Confidence: 3

39O3 Pro32.5

40Qwen3.5-122B-A10B32.3

41Qwen3.6 35B A3B31.6

42O330.4

43Qwen3.5-35B-A3B29.3

4428.2

4527.4

4625.8

4725.5

4825.0

AA-GPQA Diamond

LMSpeed rank: #32 / 74Updated: Jul 9, 2026Confidence: 3

29Qwen3.5-122B-A10B85.7

30GPT-585.4

31GLM-5 Turbo84.7

32O3 Pro84.5

32Qwen3.5-35B-A3B84.5

3484.4

3584.2

3684.1

3784.0

3882.8

AA-HLE

LMSpeed rank: #39 / 74Updated: Jul 9, 2026Confidence: 3

35Gemini 2.5 Pro21.1

36Qwen3.6 35B A3B20.2

37O320.0

38MiMo-V2-Omni19.9

39Qwen3.5-35B-A3B19.7

4018.6

4118.5

4216.2

4315.8

4414.9

AA-Omniscience Accuracy

LMSpeed rank: #56 / 68Updated: Jul 9, 2026Confidence: 3

52Llama 3.122.3

53gpt-oss-120b21.5

54Qwen3.5-27B21.0

56Qwen3.5-35B-A3B20.5

5719.7

5819.2

5918.9

6018.7

6117.5

AA-Omniscience Hallucination Rate

LMSpeed rank: #21 / 68Updated: Jul 9, 2026Confidence: 3

18GPT-5.3 Codex86.9

18Hy3 preview86.9

20Qwen3.5-122B-A10B85.5

21DeepSeek R184.0

21Qwen3.5-35B-A3B84.0

2383.7

2482.1

2582.0

2681.6

2780.5

Instruction Following score

LMSpeed rank: #26 / 66Updated: Jul 9, 2026Confidence: 3

22Qwen376.3

23GLM-576.0

24GPT-4.1 Mini75.8

25GPT-575.5

26Qwen3.5-35B-A3B75.0

2773.9

2873.2

2972.7

3072.3

3170.8

IFEval

LMSpeed rank: #9 / 14Updated: Jul 9, 2026Confidence: 3

5Qwen3.5-122B-A10B93.4

6GLM-592.6

8O192.2

9Qwen3.5-35B-A3B91.9

1090.9

1188.5

1287.4

1386.1

1483.2

AA-IFBench

LMSpeed rank: #24 / 70Updated: Jul 9, 2026Confidence: 3

19GPT-5.4 Mini73.3

21GLM-5 Turbo73.2

22GPT-573.1

24Qwen3.5-35B-A3B72.5

2572.3

2671.4

2770.4

2870.3

2970.2

110

30.4%

111Qwen3 VL 32B Instruct30.1%

112Gemini 1.5 Pro29.5%

113Qwen3.5-35B-A3B29.3%

114Olmo 3 32B Think28.6%

115Gemini 2.5 Flash Lite Preview 09-202528.5%

115Grok-228.5%

115Qwen3 30B A3B28.5%

118Nemotron 3 Nano Omni27.8%

Coding score

LMSpeed rank: #46 / 83Updated: Jul 9, 2026Confidence: 3

42GPT-564.3

43O363.6

46Qwen3.5-35B-A3B61.6

4761.5

4858.0

4956.6

5056.3

5155.3

SWE-bench Verified

LMSpeed rank: #30 / 36Updated: Jul 9, 2026Confidence: 3

26Claude Haiku 4.573.3

27Claude Sonnet 472.7

28Qwen3.5-27B72.4

29Qwen3.5-122B-A10B72.0

30Qwen3.5-35B-A3B69.2

3163.8

3254.6

3349.3

3449.0

3542.0

SWE-Rebench

LMSpeed rank: #10 / 11Updated: Jul 9, 2026Confidence: 3

2GLM-562.8

4DeepSeek V3.260.9

5Claude Sonnet 4.660.7

6Qwen3.5-27B58.9

758.7

858.5

958.2

10Qwen3.5-35B-A3B53.7

1151.9

Terminal-Bench Hard

LMSpeed rank: #46 / 69Updated: Jul 9, 2026Confidence: 3

43Qwen3.5-122B-A10B31.1

45Claude Sonnet 427.3

46Gemini 2.5 Pro26.5

46Qwen3.5-35B-A3B26.5

4825.8

4925.0

5024.2

5123.5

5220.5

AA-SciCode

LMSpeed rank: #51 / 74Updated: Jul 9, 2026Confidence: 3

47gpt-oss-120b38.9

48DeepSeek V3.238.7

49Qwen3 Max38.3

51Qwen3.5-35B-A3B37.7

5237.6

5337.3

5436.7

5536.2

5635.8

Agentic score

LMSpeed rank: #55 / 81Updated: Jul 9, 2026Confidence: 3

52Claude Haiku 4.544.1

53Qwen3.5-27B43.9

54Claude Sonnet 442.1

55Mistral Large 340.8

55Qwen3.5-35B-A3B40.8

5740.0

5839.9

5938.1

6037.5

6137.3

Terminal-Bench 2.0

LMSpeed rank: #33 / 33Updated: Jul 9, 2026Confidence: 3

25Qwen3.6 35B A3B51.5

26Kimi K2.550.8

27Claude Sonnet 4.550.0

28Qwen3.5-122B-A10B49.4

2947.1

3046.3

3141.6

3241.0

33Qwen3.5-35B-A3B40.5

BrowseComp

LMSpeed rank: #14 / 17Updated: Jul 9, 2026Confidence: 3

8Claude Opus 4.7 Max79.3

9DeepSeek V4 Flash73.2

12Qwen3.5-122B-A10B63.8

1362.0

1461.0

14Qwen3.5-35B-A3B61.0

1660.6

1752.0

OSWorld-Verified

LMSpeed rank: #14 / 16Updated: Jul 9, 2026Confidence: 3

7Claude Sonnet 4.672.1

7GPT-5.4 Mini72.1

9Claude Opus 4.566.3

10GPT-5.3 Codex64.7

11Claude Sonnet 4.561.4

1258.0

1356.2

14Qwen3.5-35B-A3B54.5

1547.3

1639.0

Tau2-Telecom

LMSpeed rank: #25 / 68Updated: Jul 9, 2026Confidence: 3

21Qwen3.5-122B-A10B93.6

22Hy3 preview92.7

23GPT-5.2 Codex92.1

24MiMo-V2-Omni91.2

25Qwen3.5-35B-A3B89.2

2688.6

2787.1

2787.1

2986.3

3086.0

Gert Labs

LMSpeed rank: #41 / 42Updated: Jul 9, 2026Confidence: 3

33Qwen3.5-27B39.4

34Gemini 3.1 Flash Lite38.5

35Grok 4.2038.4

36Hy3 preview36.9

3736.7

3830.8

3929.6

4029.6

41Qwen3.5-35B-A3B29.0

4225.6

70.0

22Claude Sonnet 467.2

22o4 Mini High67.2

22Qwen3.5-35B-A3B67.2

25Claude Opus 4.164.4

25O364.4

25O3 Pro64.4

28Qwen362.7

29Claude Haiku 4.561.6

MMLU-ProX

LMSpeed rank: #8 / 9Updated: Jul 9, 2026Confidence: 3

1Claude Opus 4.585.7

2Qwen3.6 Plus84.7

4GLM-583.1

682.2

682.2

8Qwen3.5-35B-A3B81.0

979.4

Multimodal Grounded score

LMSpeed rank: #40 / 70Updated: Jul 9, 2026Confidence: 3

36Kimi K2.562.5

37GPT-4o62.0

38Qwen3.6 35B A3B61.8

39O361.5

40Qwen3.5-35B-A3B61.1

4160.0

4259.3

4359.1

4359.1

4559.0

MMMU

LMSpeed rank: #6 / 6Updated: Jul 9, 2026Confidence: 3

1Qwen3.6 Plus86.0

2Qwen3.5-122B-A10B83.9

3Qwen3.6 27B82.9

4Qwen3.5-27B82.3

5Qwen3.6 35B A3B81.7

6Qwen3.5-35B-A3B81.4

MMVU

LMSpeed rank: #4 / 4Updated: Jul 9, 2026Confidence: 3

2Qwen3.5-122B-A10B74.7

3Qwen3.5-27B73.3

4Qwen3.5-35B-A3B72.3

MathVision

LMSpeed rank: #7 / 9Updated: Jul 9, 2026Confidence: 3

2Qwen3.6 Plus88.0

4Gemini 3 Pro86.6

5Qwen3.5-122B-A10B86.2

686.0

7Qwen3.5-35B-A3B83.9

883.0

974.3

V*

LMSpeed rank: #7 / 10Updated: Jul 9, 2026Confidence: 3

1Qwen3.6 Plus96.9

4Qwen3.6 27B94.7

5Qwen3.5-27B93.7

693.2

7Qwen3.5-35B-A3B92.7

888.0

975.9

1067.0

AA-MMMU-Pro

LMSpeed rank: #23 / 41Updated: Jul 9, 2026Confidence: 3

19Qwen3.6 27B74.6

20GPT-574.2

21GPT-5.4 Mini73.3

22GLM-5V Turbo72.8

23Qwen3.5-35B-A3B72.7

2472.5

2472.5

2472.5

2771.2

2870.6

Claude Haiku 4.5

Claude Opus 4.1

Claude Sonnet 4

Mistral Large 3

Claude Sonnet 4.5

GPT-5.1 Codex Mini

Claude Opus 4.5

Qwen3 235B A22B

Qwen3.5-122B-A10B

Qwen3.5-122B-A10B

Claude Opus 4.5

DeepSeek V4 Flash

Qwen3.6 35B A3B

Claude Opus 4.1

Claude Sonnet 4

Gemini 3.1 Flash Lite

Qwen3.6 35B A3B

Claude Opus 4.6

Claude Opus 4.6

Gemini 3.1 Flash Lite

Qwen3.6 35B A3B

Mistral Large 3

Gemini 3.1 Flash Lite

Claude Opus 4.5

Claude Opus 4.1

Claude Sonnet 4

Claude Haiku 4.5

Claude 3.5 Sonnet

Gemini 3.1 Flash Lite

DeepSeek R1 Distill Qwen

Claude Sonnet 4

Mistral Large 3

Gemini 3.1 Flash Lite

Claude 3.5 Sonnet

Gemini 2.5 Flash

Qwen3.5-122B-A10B

Claude Opus 4.7 Max

Claude Opus 4.5

Qwen3.5-122B-A10B

Claude Opus 4.5

Claude Opus 4.5

Qwen3.5-122B-A10B

Claude Opus 4.5

Claude Opus 4.6

GPT-5.1 Codex Max

Claude Opus 4.5

Claude Sonnet 4.6