How accurate is Qwen3.8 27B?

Qwen3.8 27B answered 51 of 80 questions correctly (63.8%) in the 2026-10-27 run, ranking #34 of 78 models.

How fast is Qwen3.8 27B?

Qwen3.8 27B generated a median of 49 output tokens per second in the 2026-10-27 run, ranking #60 of 78. Median time to first token was 0.5 s.

What is the latency of Qwen3.8 27B?

Qwen3.8 27B had a median time to first token of 0.5 s in the 2026-10-27 run, ranking #19 of 78 models.

How much does Qwen3.8 27B cost?

Qwen3.8 27B cost $0.0043 per question on average as billed in the 2026-10-27 run, reasoning tokens included.

What is Qwen3.8 27B API pricing?

Qwen3.8 27B is listed at $0.2 per million input tokens and $2.5 per million output tokens on OpenRouter.

What is the context window of Qwen3.8 27B?

Qwen3.8 27B accepts up to 1,000,000 tokens of context, according to OpenRouter's model catalogue.

Is Qwen3.8 27B open source?

Yes. Qwen3.8 27B is an open-weights model: Qwen publishes the weights, so you can run it yourself as well as through an API.

Who makes Qwen3.8 27B?

Qwen3.8 27B is made by Qwen. The LLM Verifications Leaderboard tested it through OpenRouter using the model ID qwen/qwen3.8-27b.

Where can I use Qwen3.8 27B via API?

Through OpenRouter, Qwen3.8 27B is served by 10 hosts: AkashML, DeepInfra, DekaLLM, Mancer 2, Modal, ModelRun, Near AI, Parasail, Reka, Wafer.

QwenOpen weightsReleased 14 Aug 2026

Qwen3.8 27B benchmarks, speed and API pricing

Reviewed by Suhas Bhairav · Tested 26 Oct 2026

Compare

Available through OpenRouter on

Tenstorrent$0.06 in · $0.5 out
Chutes$0.039 in · $1.25 out
Near AI$0.04 in · $1.35 out
Cerebras$0.99 in · $1.49 out
DeepInfra$0.15 in · $1.875 out
Ionstream$0.04 in · $1.95 out
AkashML$0.225 in · $1.98 out
Darkbloom$0.035 in · $2.2 out
Reka$0.04 in · $2.2 out
Parasail$0.24 in · $2.2 out
Modal$0.04 in · $2.3 out
Wafer$0.06 in · $2.3 out
DekaLLM$0.04 in · $2.5 out
Mancer 2$0.18 in · $2.5 out
ModelRun$0.2 in · $2.5 out
Phala$0.2 in · $2.5 out
Alibaba$0.425 in · $2.55 out
CoreWeave$0.4 in · $3 out
Novita$0.42 in · $3 out
Cloudflare$0.45 in · $3.2 out
Venice$0.45 in · $3.2 out

Intelligence63.8% correct#34 of 72

Speed49tok/s#60 of 72

Cost$0.0043per question#47 of 72List $0.2 in · $2.5 out per 1M

Verbosity1,913tokens per question#48 of 72

Summary

Answers 51 of 80 verification questions correctly (63.8%), placing #34 of 72 models tested on the current suite.
Generates 49 output tokens per second, faster than 17% of the models the leaderboard measures.
Starts responding after 0.5 s (median time to first token), including thinking time.
Costs $0.0043 per question as billed, #29 cheapest of 29 models priced $0.15 to $1 per 1M tokens.

Specifications

Reasoning: Yes
Input: text, image, video
Output: text
Context window: 1M tokens
Weights: Open weights
Knowledge cutoff: —
OpenRouter ID: qwen/qwen3.8-27b
Last tested: 26 Oct 2026

Peer group: 31 models with a blended list price $0.15 to $1 per 1M tokens (3 input : 1 output). Charts below compare Qwen3.8 27B with all models; cost and speed charts focus on this peer group.

Intelligence

Verification score: the share of 80 questions from open datasets answered correctly. See the benchmark datasets.

Score vs. cost per question

Qwen3.8 27B circled; dashed line joins models no cheaper model beats

Score vs. end-to-end response time

Qwen3.8 27B circled · accurate and quick sits top left

Score vs. tokens per question

Qwen3.8 27B circled · output tokens, reasoning included

Verification score

All models tested on the current suite

61 more models · see the full leaderboard

6 of 78 tested models have no verification score

LFM2.5-2.6B: 80 of 80 questions failed on the host's side
Ling 3.0 Flash VL: not yet run on the current question suite
Qwen3.5-122B-A10B: not yet run on the current question suite
Qwen3 Next 80B A3B Thinking: not yet run on the current question suite
Nova Premier 1.0: not yet run on the current question suite
GPT-5.5 Pro: not yet run on the current question suite

Cost

Average cost of one question, split into input, reasoning and answer tokens at each model's list price. Reasoning tokens are billed as output.

Cost per question, by token type

Peer group: blended price $0.15 to $1

1Llama 4 Scout$0.00025
2GPT-6 Luna$0.00025
3Ministral 3 8B 2512$0.00027
4Llama 3.3 70B Instruct$0.00031
5Ministral 3 14B 2512$0.00031
6GLM 5.3 Flash$0.00033
7gpt-oss-120b$0.00038
8Inkling Small$0.00047
9Claude Haiku 5.5$0.00048
10MiMo-V2.6-Flash$0.00056
11Llama 4 Maverick$0.00057
12GLM 5.3$0.00076
13DeepSeek V4.1 Flash 1017$0.00084
14MiniMax M3.1 Flash$0.0009
15DeepSeek V4 Flash Vision Exp$0.00092
16Hy3$0.0013
17Gemma 4 31B$0.0014
18Nemotron 3.5 Super VL$0.0014
19Qwen3 Coder Next$0.0017
20Solar Pro 4$0.0018
21Mistral Small 4$0.0019
22MiMo-V2.6-Pro$0.002
23Trinity Large Thinking$0.0025
24MiniMax M3$0.0027
25LongCat 2.0$0.0032
26Qwen3.7 Plus$0.0033
27Gemini 3.5 Flash Lite$0.0034
28Qwen3.6 35B A3B$0.0034
29Qwen3.8 27B$0.0049

Input Reasoning Answer

Cost per question, as billed

All models · lower is better

1Granite 4.0 Micro$0.000051
2gpt-oss-20b$0.000066
3Granite 4.2 8B$0.000077
4Nova Micro 1.0$0.000087
5Mercury 2.5$0.00011
6Phi 4$0.00012
7gpt-oss-120b$0.00013
8GLM 5.3 Flash$0.00015
9Ministral 3 3B 2512$0.00019
10GPT-6 Luna$0.00025
47Qwen3.8 27B$0.0043

61 more models · see the full leaderboard

6 of 78 tested models have no cost per question

LFM2.5-2.6B: no verification score (80 of 80 questions failed on the host's side)
Ling 3.0 Flash VL: no verification score (not yet run on the current question suite)
Qwen3.5-122B-A10B: no verification score (not yet run on the current question suite)
Qwen3 Next 80B A3B Thinking: no verification score (not yet run on the current question suite)
Nova Premier 1.0: no verification score (not yet run on the current question suite)
GPT-5.5 Pro: no verification score (not yet run on the current question suite)

Speed and latency

Median of 3 streamed runs of the same prompt, measured 25 Oct 2026.

Output speed

Peer group ($0.15 to $1) · higher is better

16 more models · see the full leaderboard

2 of 31 tested models have no output speed

LFM2.5-2.6B: speed not measured
Qwen3 Next 80B A3B Thinking: speed runs failed on the host's side

Time to first token

Peer group ($0.15 to $1) · lower is better

17 more models · see the full leaderboard

2 of 31 tested models have no time to first token

LFM2.5-2.6B: speed not measured
Qwen3 Next 80B A3B Thinking: speed runs failed on the host's side

End-to-end response time

Peer group ($0.15 to $1) · lower is better

16 more models · see the full leaderboard

2 of 31 tested models have no end-to-end response time

LFM2.5-2.6B: speed not measured
Qwen3 Next 80B A3B Thinking: speed runs failed on the host's side

Latency vs. output speed

Qwen3.8 27B circled · quick to start and fast sits top left

Results by category

Questions answered correctly in each category

Science21/30

Engineering6/15

Medicine5/11

Economics8/9

Law3/6

Humanities & social science8/9

More from Qwen

Qwen#15Qwen3.8 Max (0902)75.0% verification score46 tok/s1.2 s first token$0.011 / question

Qwen#36Qwen3 Coder Next62.5% verification score106 tok/s0.4 s first token$0.0017 / question

Qwen#38Qwen3.7 Plus60.0% verification score62 tok/s1.1 s first token$0.0033 / question

All Qwen models →

Compare Qwen3.8 27B

Head to head with its closest rivals, other Qwen models and the top models overall.

Hosts and prices

Every host serving this model through OpenRouter, from OpenRouter's catalogue

Host	Input / 1M	Output / 1M	Context	Max output	Quantization
Tenstorrent	$0.06	$0.5	262k	66k	fp4
Chutes	$0.039	$1.25	262k	66k	fp8
Near AI	$0.04	$1.35	262k	66k	fp8
Cerebras	$0.99	$1.49	66k	33k	fp16
DeepInfra	$0.15	$1.875	262k	236k	bf16
Ionstream	$0.04	$1.95	262k	66k	fp8
AkashML	$0.225	$1.98	262k	131k	fp8
Darkbloom	$0.035	$2.2	262k	33k	fp4
Reka	$0.04	$2.2	262k	236k	fp8
Parasail	$0.24	$2.2	262k	236k	fp8
Modal	$0.04	$2.3	262k	236k	—
Wafer	$0.06	$2.3	262k	236k	fp8
DekaLLM	$0.04	$2.5	262k	236k	—
Mancer 2	$0.18	$2.5	262k	236k	fp8
ModelRun	$0.2	$2.5	262k	236k	fp4
Phala	$0.2	$2.5	1M	262k	—
Alibaba	$0.425	$2.55	1M	131k	—
CoreWeave	$0.4	$3	262k	236k	fp8
Novita	$0.42	$3	1M	131k	—
Cloudflare	$0.45	$3.2	262k	236k	—
Venice	$0.45	$3.2	262k	66k	fp8

Sources

Scores, speed and billed cost are the leaderboard's own measurements. Prices, context window and hosts come from OpenRouter's catalogue on the test date.

Qwen3.8 27B on OpenRouter: list prices and hosts
Model card on Hugging Face: weights and licence
SuperGPQA (paper): 50 of the questions, ODC-BY 1.0
MMLU-Pro (paper): 30 of the questions, MIT
OpenRouter generation records: where billed cost and native token counts come from
Methodology and editorial standards

Qwen3.8 27B: frequently asked questions

How accurate is Qwen3.8 27B?: Qwen3.8 27B answered 51 of 80 questions correctly (63.8%) in the 2026-10-27 run, ranking #34 of 78 models.
How fast is Qwen3.8 27B?: Qwen3.8 27B generated a median of 49 output tokens per second in the 2026-10-27 run, ranking #60 of 78. Median time to first token was 0.5 s.
What is the latency of Qwen3.8 27B?: Qwen3.8 27B had a median time to first token of 0.5 s in the 2026-10-27 run, ranking #19 of 78 models.
How much does Qwen3.8 27B cost?: Qwen3.8 27B cost $0.0043 per question on average as billed in the 2026-10-27 run, reasoning tokens included.
What is Qwen3.8 27B API pricing?: Qwen3.8 27B is listed at $0.2 per million input tokens and $2.5 per million output tokens on OpenRouter.
What is the context window of Qwen3.8 27B?: Qwen3.8 27B accepts up to 1,000,000 tokens of context, according to OpenRouter's model catalogue.
Is Qwen3.8 27B open source?: Yes. Qwen3.8 27B is an open-weights model: Qwen publishes the weights, so you can run it yourself as well as through an API.
Who makes Qwen3.8 27B?: Qwen3.8 27B is made by Qwen. The LLM Verifications Leaderboard tested it through OpenRouter using the model ID qwen/qwen3.8-27b.
Where can I use Qwen3.8 27B via API?: Through OpenRouter, Qwen3.8 27B is served by 10 hosts: AkashML, DeepInfra, DekaLLM, Mancer 2, Modal, ModelRun, Near AI, Parasail, Reka, Wafer.