Released on Mar 10, 2026
Qwen3.5 API Benchmarks, Pricing & Provider Data
Compare Qwen3.5 with another model
Choose a model to open its comparison page.
LLM
Qwen3.5 benchmark, API pricing, and provider data cover 331 API providers, with prices starting at $0.0050/M. Qwen3.5 free API options are available from 9 providers. The page also shows measured API speed and first-token latency.
Alibaba Qwen3.5 is a Qwen3 generation model with improved reasoning, multilingual support, and efficient inference for chat, coding, and agent applications.
Quality
#59of 101
54.0
LMSpeed score
Speed
#78of 78
51char/s
10.34 s
Cost
#1of 182
$0.0050/ 1M · 8:1 in:out
$0.0008 in · $0.0042 out
Category Performance
Observed-capability estimates with uncertainty reported separately from independent benchmark families.
- Coverage
- 8 / 8
- Methodology
- V3.0
#1Multilingual54.6Estimated2/4 Measured dimensions
80% interval: 42.4–66.8
cross language understanding54.9
nova · nova63
multilingual generationPrior only
reasoning transfer54.2
mmlu_prox · mmlu_pro_x
low resource robustnessPrior only
#2Instruction following53.7Estimated2/4 Measured dimensions
80% interval: 41.8–65.6
constraint following51.5
ifeval · ifeval
structured outputPrior only
novel instruction generalization55.8
ifbench · aa_if_bench
long multiturn instructionPrior only
#3Knowledge53.1Estimated2/4 Measured dimensions
80% interval: 41.4–64.7
broad knowledge57.1
c_eval · c_eval / mmlu_redux · mmlu_redux
professional knowledge49.1
supergpqa · super_gpqa
factualityPrior only
retrieval open bookPrior only
#4Reasoning52.3RatedGlobal rank #353/4 Measured dimensions
80% interval: 44.1–60.4
abstract logicPrior only
scientific causal41.1
critpt · critpt / gpqa · gpqa / hle · hle
multistep constraints59.4
mmlu_pro · mmlu_pro
evidence verification56.3
ai_needle · ai_needle / lcr · lcr / longbench · long_bench_v2
#5Multimodal50.2RatedGlobal rank #64/4 Measured dimensions
80% interval: 43.7–56.6
perception ocr57.3
v_star · v_star
document spatial48.6
charxiv · charxiv
visual reasoning52.5
mathvision · math_vision / mmmu_pro · mmmu_pro
video action42.3
screenspot · screen_spot_pro / video_mmmu · video_mmmu
#6Agents44.6RatedGlobal rank #504/4 Measured dimensions
80% interval: 39.3–49.8
planning48.8
deep_planning · gert_labs
tool use41.9
mcp_atlas · mcp_atlas / mcp_tasks · mcp_tasks / tau · tau2_bench / tau · tau3_bench / toolathlon · toolathlon
environment execution45.2
browsecomp · browse_comp / gdpval_aa · benchlm_agentic_gdpval_aa / terminalbench · benchlm_agentic_terminal_bench2 / vita_bench · vita_bench / wide_research · wide_research
recovery reliability42.5
apex_agents · apex_agents_aa / claw_eval · claw_eval / researchclaw · research_claw_bench
#7Math41.4Estimated2/4 Measured dimensions
80% interval: 30.0–52.9
foundational mathPrior only
competition math41.2
aime · aime2026 / hmmt · hmmt_feb2025 / hmmt · hmmt_feb2026 / hmmt · hmmt_nov2025
proof frontierPrior only
applied tool math41.7
mmanswer · mm_answer_bench
#8Coding40.3RatedGlobal rank #373/4 Measured dimensions
80% interval: 31.4–49.3
code generation30.4
livecodebench · live_code_bench_v6 / scicode · scicode
repository engineering41.3
swe_pro · swe_pro
debugging testing49.3
swe_verified · swe_verified
tooling qualityPrior only
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
262.1Ktokens
8K128K1M4M
262.1K
Features
Technical Details
- Input
- Output
- Total parameters
- 9B
- Released
- Mar 2026
- 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
Rankings
Excels at
It's decent at
Falls behind in
Detailed scores
Updated: Sep 2, 2026Overall
undefined metric} other undefined metrics}}
Overall score54.0#59 / 101
Overall
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Speed & latency
undefined metric} other undefined metrics}}
Output speed20.7 tok/s#78 / 78Time to first token0.45 s#7 / 78
Speed & latency
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Time to first token
LMSpeed rank#7
Score0.45 s
Sourceartificialanalysis.ai
UpdatedSep 2, 2026
Confidence4
Pricing
undefined metric} other undefined metrics}}
Input price$0.030/M#1 / 182Output price$0.150/M#2 / 182
Pricing
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Score44.6#5080% interval39.3–49.84/4 Measured dimensionsAgentic score32.4#60 / 70Terminal-Bench 2.052.5#42 / 53
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Planning & decomposition48.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
deep_planning · gert_labs · z -0.14 · q 1.00
Tool use41.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mcp_atlas · mcp_atlas · z -2.04 · q 1.00
mcp_tasks · mcp_tasks · z 0.27 · q 0.63
tau · tau2_bench · z 0.69 · q 1.00
tau · tau3_bench · z -0.70 · q 1.00
toolathlon · toolathlon · z -1.22 · q 1.00
Environment & long-horizon execution45.2
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
browsecomp · browse_comp · z -1.27 · q 1.00
gdpval_aa · benchlm_agentic_gdpval_aa · z -0.55 · q 1.00
terminalbench · benchlm_agentic_terminal_bench2 · z -0.69 · q 1.00
vita_bench · vita_bench · z 0.37 · q 1.00
wide_research · wide_research · z 0.00 · q 1.00
Recovery & completion reliability42.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
apex_agents · apex_agents_aa · z -0.76 · q 1.00
claw_eval · claw_eval · z -0.45 · q 1.00
researchclaw · research_claw_bench · z -1.29 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Coding
V3.0undefined metric} other undefined metrics}} · Rated
Score40.3#3780% interval31.4–49.33/4 Measured dimensionsSciCode16.1%#204 / 207Coding score46.0#63 / 82
Coding
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Code generation30.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
livecodebench · live_code_bench_v6 · z -0.71 · q 0.75
scicode · scicode · z -3.00 · q 1.00
Repository engineering41.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
swe_pro · swe_pro · z -1.08 · q 1.00
Debugging & testing49.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
swe_verified · swe_verified · z -0.12 · q 1.00
Tool-assisted development & qualityPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
SciCodeV3 evidence
LMSpeed rank#204
Score16.1%
Sourceartificialanalysis.ai
UpdatedSep 2, 2026
Confidence4
Reasoning
V3.0undefined metric} other undefined metrics}} · Rated
Score52.3#3580% interval44.1–60.43/4 Measured dimensionsMMLU-Pro87.8%#8 / 129GPQA77.1%#109 / 214
Reasoning
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Abstract logicPrior only
Scientific & causal reasoning41.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
critpt · critpt · z -0.21 · q 1.00
gpqa · gpqa · z -1.33 · q 1.00
hle · hle · z -1.17 · q 1.00
Multi-step constraints59.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mmlu_pro · mmlu_pro · z 1.05 · q 1.00
Evidence integration & verification56.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ai_needle · ai_needle · z 0.09 · q 0.50
lcr · lcr · z 0.16 · q 1.00
longbench · long_bench_v2 · z 1.14 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Knowledge
V3.0undefined metric} other undefined metrics}} · Estimated
Score53.180% interval41.4–64.72/4 Measured dimensionsKnowledge score53.5#57 / 70SuperGPQA70.4#8 / 16
Knowledge
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Broad knowledge57.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
c_eval · c_eval · z 0.71 · q 0.63
mmlu_redux · mmlu_redux · z 0.56 · q 0.75
Professional knowledge49.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
supergpqa · super_gpqa · z 0.13 · q 1.00
FactualityPrior only
Retrieval & open-book usePrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Artificial Analysis Intelligence Index
LMSpeed rank#66
Score34.3
Sourcebenchlm.ai
UpdatedSep 2, 2026
Confidence4
AA-Omniscience Hallucination Rate
LMSpeed rank#23
Score88.9
Sourcebenchlm.ai
UpdatedSep 2, 2026
Confidence4
Math
V3.0undefined metric} other undefined metrics}} · Estimated
Score41.480% interval30.0–52.92/4 Measured dimensionsMath score74.3#14 / 62AIME2693.3#12 / 14
Math
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Foundational mathPrior only
Competition math41.2
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
aime · aime2026 · z -1.54 · q 1.00
hmmt · hmmt_feb2025 · z 0.00 · q 0.88
hmmt · hmmt_feb2026 · z 0.07 · q 1.00
hmmt · hmmt_nov2025 · z -0.25 · q 1.00
Advanced proofsPrior only
Applied & tool-assisted math41.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mmanswer · mm_answer_bench · z -1.06 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Multilingual
V3.0undefined metric} other undefined metrics}} · Estimated
Score54.680% interval42.4–66.82/4 Measured dimensionsMultilingual score69.7#4 / 11MMLU-ProX84.7#4 / 11
Multilingual
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Cross-language understanding54.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
nova · nova63 · z 0.64 · q 0.88
Multilingual generationPrior only
Reasoning transfer54.2
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mmlu_prox · mmlu_pro_x · z 0.77 · q 1.00
Low-resource robustnessPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Multimodal
V3.0undefined metric} other undefined metrics}} · Rated
Score50.2#680% interval43.7–56.64/4 Measured dimensionsMultimodal Grounded score60.9#27 / 44MMMU-Pro79.0#13 / 31
Multimodal
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Perception & OCR57.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
v_star · v_star · z 0.72 · q 1.00
Document & spatial understanding48.6
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
charxiv · charxiv · z -0.07 · q 1.00
Visual reasoning52.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mathvision · math_vision · z 0.40 · q 1.00
mmmu_pro · mmmu_pro · z 0.12 · q 1.00
Video & grounded action42.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
screenspot · screen_spot_pro · z -1.12 · q 1.00
video_mmmu · video_mmmu · z 0.03 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Instruction following
V3.0undefined metric} other undefined metrics}} · Estimated
Score53.780% interval41.8–65.62/4 Measured dimensionsInstruction Following score87.4#12 / 25IFEval92.6#8 / 16
Instruction following
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Constraint following51.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ifeval · ifeval · z 0.00 · q 1.00
Structured outputPrior only
Novel-instruction generalization55.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ifbench · aa_if_bench · z 0.50 · q 1.00
Multi-turn & long instructionsPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Pricing Comparison
Compare Qwen3.5 API pricing across 322 providers. Prices range from $0.0050/M to $1718.28/M. RenRen API offers the lowest rate at $0.0050/M. 9 providers offer free API credits or a free tier.
| Provider | Health | Model Variant | Group | Input ($/M) | Output ($/M) | Speed (t/s) | First token | Audit |
|---|---|---|---|---|---|---|---|---|
L1 100% | qwen3.5-35b-a3b | default | -9%$0.027/M | $0.219/M | — | — | — | |
L1 100% | qwen3.5-27b | default | $0.041/M | $0.329/M | — | — | — | |
L1 100% | qwen3.5-122b-a10b | default | $0.055/M | $0.438/M | — | — | — | |
L1 100% | qwen3.5-397b-a17b | default | $0.082/M | $0.493/M | — | — | — | |
L1 100% | Qwen3.5-35B-A3B | default | $0.020/request | - | — | — | — | |
L1 100% | Qwen3.5-122B-A10B | default | $0.020/request | - | — | — | — | |
L1 100% | qwen3.5-35b-a3b | qwen | $0.160/M | $1.28/M | — | — | — | |
L1 100% | qwen3.5-27b | qwen | $0.240/M | $1.92/M | — | — | — | |
L1 100% | qwen3.5-122b-a10b | qwen | $0.320/M | $2.56/M | — | — | — | |
L1 100% | qwen3.5-397b-a17b | qwen | $0.480/M | $2.88/M | — | — | — | |
L1 100% | qwen3.5-35b-a3b | default | $0.055/M | $0.438/M | — | — | — | |
L1 100% | qwen3.5-27b | default | $0.082/M | $0.658/M | — | — | — | |
L1 100% | qwen3.5-122b-a10b | default | $0.110/M | $0.877/M | — | — | — | |
L1 100% | qwen3.5-397b-a17b | default | $0.164/M | $0.986/M | — | — | — | |
L1 100% | qwen3.5-397b-a17b | default | $10.27/M | $10.27/M | — | — | — | |
L1 100% | qwen3.5-35b-a3b | default | -9%$0.027/M | $0.219/M | — | — | — | |
L1 100% | qwen3.5-27b | default | $0.041/M | $0.329/M | — | — | — | |
L1 100% | qwen3.5-122b-a10b | default | $0.055/M | $0.438/M | — | — | — | |
L1 100% | qwen3.5-397b-a17b | default | $0.082/M | $0.493/M | — | — | — | |
L1 100% L2 62% | qwen3.5-122b-a10b | nvidia | $0.400/M | $3.20/M | — | — | — |
Showing 20 model IDs of 131.
Alternatives & Similar Models
DeepSeek V4 Flash
deepseek-v4-flash
DeepSeek V4 Flash is a fast, cost-efficient language model in the DeepSeek V4 family, optimized for low-latency chat, coding assistance, and high-throughput API workloads while retaining strong reasoning quality.
119 shared providers
GLM-5.1
glm-5-1
Zhipu GLM-5.1 is a next-generation GLM model aimed at frontier reasoning, coding, and bilingual agent applications.
117 shared providers
Kimi K2.5
kimi-k2-5
Moonshot Kimi K2.5 is an open-weight multimodal agent model with native vision and text input, strong coding performance, and a 256K context window.
116 shared providers
DeepSeek V4 Pro
deepseek-v4-pro
DeepSeek V4 Pro is the professional-tier DeepSeek V4 model, targeting frontier reasoning, coding, and agent workflows with maximum capability.
114 shared providers
MiniMax M2.7
minimax-m2-7
MiniMax M2.7 is a high-tier M2-series model tuned for complex reasoning, long-context dialogue, and production-grade API workloads.
112 shared providers
MiniMax M2.5
minimax-m2-5
MiniMax M2.5 is MiniMax's flagship text model for coding and agents, with SOTA-level programming and agentic performance, improved token efficiency, and fast high-TPS API deployment.
111 shared providers
Frequently Asked Questions
- What benchmark data does Qwen3.5 include?
- LMSpeed shows Qwen3.5 benchmark context, API price, output speed, first-token latency, and provider data across 331 providers when those signals are available.
- What is the Qwen3.5 API price?
- Qwen3.5 has pricing from 331 providers, ranging from $0.0050/M to $1718.28/M. RenRen API has the lowest listed price.
- What does the Qwen3.5 API pricing table include?
- The Qwen3.5 API pricing table compares 331 providers by input price, output price, free tier status, speed, first-token latency, and recent health data when available.
- Which provider has the cheapest Qwen3.5 API pricing?
- RenRen API currently has the lowest listed Qwen3.5 price at $0.0050/M across 331 providers.
- Can I compare Qwen3.5 API price and speed together?
- Yes. LMSpeed shows Qwen3.5 API price, output speed, first-token latency, and provider health on the same page so you can compare cost and performance together.
- Is Qwen3.5 API free?
- Yes, Qwen3.5 free API options are available through 9 providerundefined other undefined} on LMSpeed, including Dext API, ApiToken Online, Dext API, Zero API, Zero API. These providers offer free API credits or a free tier with no per-token charges.
- Where can I get Qwen3.5 free API access?
- LMSpeed currently lists 9 free API providerundefined other undefined} for Qwen3.5: Dext API, ApiToken Online, Dext API, Zero API, Zero API. Check each provider row before using it because free tier limits can change.