Qwen3.8 27B vs gpt-oss-20b: Release Comparison

Name: Intelligence
Creator: Artificial Analysis
License: https://artificialanalysis.ai/docs/legal/Terms-of-Use.pdf

Comparison of the Qwen3.8 27B and gpt-oss-20b releases: 4 and 2 models respectively, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across all 6 models.

For intelligence, Qwen3.8 27B scores highest: Qwen3.8 27B (Xhigh) at 34, against gpt-oss-20b (Low) at 10 for gpt-oss-20b.
For output speed, gpt-oss-20b is fastest: gpt-oss-20b (Low) at 296 t/s, against Qwen3.8 27B (Low) at 53 t/s for Qwen3.8 27B.
For cost per task, gpt-oss-20b is cheapest: gpt-oss-20b (High) at $0.01, against Qwen3.8 27B (Xhigh) at $1.01 for Qwen3.8 27B.

Model Playground

Releases compared


	Qwen3.8 27B (xhigh)	Qwen3.8 27B (medium)	Qwen3.8 27B (low)	Qwen3.8 27B (non-reasoning)	gpt-oss-20b (low)	gpt-oss-20b (high)
Intelligence Index	34	28	26	20	10*	9
Finance & Accounting	34	28	27	19	--	7
Strategy & Ops	37	34	29	16	--	5
Legal	35	27	28	22	--	7
Healthcare & Medical	34	--	--	--	--	7
Engineering	33	25	24	22	--	12
Economics	43	31	32	30	--	11
AA-Briefcase v1.1	1397	1377	1273	898	--	0
GDPval-AA v2.1	1419	1381	1362	1072	--	339
AutomationBench-AA	48.2%	40.3%	30.1%	12.4%	--	0.2%
Terminal-Bench 4.0	5.6%	5.1%	2.5%	0.0%	--	0.0%
SciCode	46.6%	39.0%	40.0%	36.2%	--	38.9%
Humanity's Last Exam	33.9%	14.1%	14.0%	12.1%	5.3%	11.0%
GDP.pdf	16.6%	12.8%	10.4%	5.2%	--	2.0%
CritPt	5.4%	0.0%	0.0%	0.3%	0.0%	1.4%
AA-Omniscience	−10	−36	−27	−8	−59	−63
AA-LCR v1.1	82.0%	79.7%	77.3%	69.3%	31.0%	34.7%

Cost per TaskUSD	$1.01	$1.13	$1.05	$2.49	--	$0.01
Input PriceUSD/1M Tokens	$0.50	$0.50	$0.50	$0.50	$0.07	$0.07
Output PriceUSD/1M Tokens	$3.00	$3.00	$3.00	$3.00	$0.215	$0.18
Cache Hit PriceUSD/1M Tokens	$0.10	$0.10	$0.10	$0.10	--	--
Cache Write PriceUSD/1M Tokens	--	--	--	--	--	--

Output SpeedTokens/s	46	50	53	51	296	223
Time to First Tokens	3.80	3.92	3.78	3.88	0.74	0.73
Time to First Answer Tokens	47.16	43.77	41.73	3.88	7.50	9.69
End-to-End Response Times	58.00	53.73	51.22	13.62	9.19	11.92
Time per Tasks	1,087.00	968.02	820.57	556.01	--	65.16

Context Window	256k	256k	256k	256k	131k	131k
Creator	Alibaba	Alibaba	Alibaba	Alibaba	OpenAI	OpenAI
License	Open	Open	Open	Open	Open	Open
Input Modality	Supports: text, image, and video	Supports: text, image, and video	Supports: text, image, and video	Supports: text, image, and video	Supports: text	Supports: text
Output Modality	Supports: text	Supports: text	Supports: text	Supports: text	Supports: text	Supports: text
Providers	+6			+3	+2	+3

* Estimated

Intelligence

Artificial Analysis Intelligence Index · Higher is better

Speed

Output tokens per second · Higher is better

Cost per Task

Weighted average cost (USD) per Intelligence Index task · Lower is better

Intelligence

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.3.2 incorporates 10 evaluations: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1

Estimate (independent evaluation forthcoming)

Artificial Analysis Intelligence Index v4.3.2 includes: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

Intelligence Index vs. Cost per Intelligence Index Task

Artificial Analysis Intelligence Index · Weighted average cost (USD) per Artificial Analysis Intelligence Index task

Most attractive quadrant

Pareto line

AA-Briefcase v1.1

GDPval-AA v2.1

Cost per Task (USD, Log Scale)

Weighted average cost per Intelligence Index task. Each evaluation’s cost is calculated from input, cache hit, cache write, reasoning, and answer token prices, divided by task count, and weighted by its Intelligence Index weight.

Cost

Cost per Intelligence Index Task

Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better

Speed & Latency

Output Speed

Output tokens per second · Higher is better

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Figures represent performance of the model's first-party API or the median across providers where a first-party API is not available.

Capability Scores

Capability Indexes

Measures the performance of models on specific capabilities and industries

Finance & Accounting Index

Incorporates 7 evaluations: AA-Omniscience, GDPval-AA v2.1, AA-Briefcase v1.1, Humanity's Last Exam, AutomationBench-AA, AA-LCR v1.1, GDP.pdf · Higher is better

Strategy & Ops Index

Incorporates 6 evaluations: GDPval-AA v2.1, AA-Briefcase v1.1, AA-Omniscience, AutomationBench-AA, AA-LCR v1.1, GDP.pdf · Higher is better

Legal Index

Incorporates 7 evaluations: AA-Omniscience, GDPval-AA v2.1, AA-Briefcase v1.1, Humanity's Last Exam, AA-LCR v1.1, GDP.pdf, AutomationBench-AA · Higher is better

Healthcare & Medical Index

Incorporates 6 evaluations: AA-Omniscience, GDPval-AA v2.1, AA-Briefcase v1.1, MLCR-AA, Humanity's Last Exam, AutomationBench-AA · Higher is better

Engineering Index

Incorporates 6 evaluations: AA-Omniscience, Humanity's Last Exam, CritPt, GDPval-AA v2.1, AA-Briefcase v1.1, Terminal-Bench 4.0 · Higher is better

Economics Index

Incorporates 5 evaluations: AA-Omniscience, Humanity's Last Exam, GDPval-AA v2.1, AA-Briefcase v1.1, AA-LCR v1.1 · Higher is better

Qwen3.8 27B vs gpt-oss-20b: Release Comparison

Intelligence

Speed

Cost per Task

Intelligence

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index

Intelligence Index vs. Cost per Intelligence Index Task

Cost per Intelligence Index Task

Artificial Analysis Intelligence Index

Cost

Cost per Intelligence Index Task

Cost per Intelligence Index Task

Speed & Latency

Output Speed

Output Speed

Model Performance Representation

Capability Scores

Capability Indexes