GPT-6 Luna vs Qwen3.8-27B

Updated November 6, 2026. Rank says GPT-6 Luna is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.

Decision reading

GPT-6 Luna has the higher public score estimate, 62.59 versus 54.62, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 5 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Model A

GPT-6 Luna

OpenAI

62.59/100

Supported · Public rank #43

90% interval 52.0–73.2

Model B

Qwen3.8-27B

Alibaba

54.62/100

Supported · Public rank #74

90% interval 42.0–67.3

Shared results: 5
GPT-6 Luna only: 18
Qwen3.8-27B only: 32
Like-for-like categories: 2 / 8

Supported: GPT-6 Luna and Qwen3.8-27BHow the comparison works

Which one for your work

Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.

Coding work
Code generation, repair, and software-engineering tasks
GPT-6 Luna
GPT-6 Luna has the higher public coding point estimate, 51.8 to 46.1, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
Confidence: limited
Long documents
Prompts that approach the documented context limit
GPT-6 Luna
GPT-6 Luna has the larger documented context window.
Confidence: documented

Show secondary and unsupported calls

Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
GPT-6 Luna is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates

Which one for a specific job

Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.

Pick the job you would use them for

The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.

51.8GPT-6 Luna46.1Qwen3.8-27B

Like-for-like · BenchAlign v5.8

GPT-6 Luna has the higher coding point estimate. Conditional score ranges do not establish rank confidence.

Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.

Open the coding leaderboard Shortlist models for this job

Same basis rules as the category table below

Shape of the matched evidence

Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.

A shared-evidence shape is not available.

BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign v5.8 lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.

Coding

Like-for-like

GPT-6 Luna: 51.8; Supported · #42/157
Qwen3.8-27B: 46.1; Supported · #57/157
Basis: BenchAlign v5.8 lane · 3 vs 9 public rows
Reading: GPT-6 Luna leads · intervals overlap

Knowledge

Like-for-like

GPT-6 Luna: 63.3; Supported · #40/178
Qwen3.8-27B: 48.7; Supported · #86/178
Basis: BenchAlign v5.8 lane · 5 vs 6 public rows
Reading: GPT-6 Luna leads · intervals overlap

Agentic

Directional only

GPT-6 Luna: 53.5; Estimated · #45/131
Qwen3.8-27B: 62.6; Supported · #20/131
Basis: BenchAlign v5.8 lane · 5 vs 9 public rows
Reading: Directional only

Reasoning

Directional only

GPT-6 Luna: 57.5; #22/25
Qwen3.8-27B: 79.6; #7/25
Basis: Provisional lane · 2 vs 0 weighted rows
Reading: Directional only

Multimodal

Not comparable

GPT-6 Luna: 81.9; Unranked · 1 rankable row
Qwen3.8-27B: 82.3; #17/59
Basis: Provisional lane · 0 vs 1 weighted rows
Reading: Not comparable

Multilingual

Not comparable

GPT-6 Luna: Not ranked
Qwen3.8-27B: Not ranked
Basis: Provisional lane · 0 vs 0 weighted rows
Reading: Not comparable

Instruction following

Not comparable

GPT-6 Luna: Not ranked
Qwen3.8-27B: 86.9; #44/157
Basis: Provisional lane · 0 vs 1 weighted rows
Reading: Not comparable

Math

Not comparable

GPT-6 Luna: 80.2; Unranked · 2 rankable rows
Qwen3.8-27B: Not ranked
Basis: Provisional lane · 2 vs 0 weighted rows
Reading: Not comparable

Category scores with the server-provided evidence basis for GPT-6 Luna and Qwen3.8-27B
Category	GPT-6 Luna	Qwen3.8-27B	Basis	Reading
Coding	51.8Supported · #42/157	46.1Supported · #57/157	Like-for-likeBenchAlign v5.8 lane · 3 vs 9 public rows	GPT-6 Luna leads · intervals overlap
Knowledge	63.3Supported · #40/178	48.7Supported · #86/178	Like-for-likeBenchAlign v5.8 lane · 5 vs 6 public rows	GPT-6 Luna leads · intervals overlap
Agentic	53.5Estimated · #45/131	62.6Supported · #20/131	Directional onlyBenchAlign v5.8 lane · 5 vs 9 public rows	Directional only
Reasoning	57.5#22/25	79.6#7/25	Directional onlyProvisional lane · 2 vs 0 weighted rows	Directional only
Multimodal	81.9Unranked · 1 rankable row	82.3#17/59	Not comparableProvisional lane · 0 vs 1 weighted rows	Not comparable
Multilingual	Not ranked	Not ranked	Not comparableProvisional lane · 0 vs 0 weighted rows	Not comparable
Instruction following	Not ranked	86.9#44/157	Not comparableProvisional lane · 0 vs 1 weighted rows	Not comparable
Math	80.2Unranked · 2 rankable rows	Not ranked	Not comparableProvisional lane · 2 vs 0 weighted rows	Not comparable

Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.

Supported evidence per lane · bars run 0–100Methodology

What each workload costs

Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.

Chat turn

1K fresh input + 500 output tokens

GPT-6 Luna: $0.00035; Fits in one request
Qwen3.8-27B: Self-hosted; infrastructure cost varies; Fits in one request

Qwen3.8-27B has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

GPT-6 Luna: $0.0065; Fits in one request
Qwen3.8-27B: Self-hosted; infrastructure cost varies; Fits in one request

Qwen3.8-27B has no comparable published API token rate.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

GPT-6 Luna: $0.009; Fits in one request
Qwen3.8-27B: Self-hosted; infrastructure cost varies; Fits in one request; Cached-input rate unavailable

Qwen3.8-27B has no comparable published API token rate.

Cached input falls back to the list input rate only where a cached rate is unpublished

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

SpecificationGPT-6 LunaQwen3.8-27B

Context window

Maximum documented context; output-token limits may be lower.

GPT-6 Luna

1.05M

OpenAI GPT-6 Luna model documentation

Qwen3.8-27B

262K

Qwen3.8-27B model card

API model ID

GPT-6 Luna

gpt-6-luna

OpenAI GPT-6 Luna model documentation

Qwen3.8-27B

Not sourced

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

GPT-6 Luna

$0.01 per 1M cached input tokens

OpenAI GPT-6 Luna model documentation

Qwen3.8-27B

No comparable hosted API rate

Qwen3.8-27B model card

Documented inputs

GPT-6 Luna

text, image

OpenAI GPT-6 Luna model documentation

Qwen3.8-27B

Not sourced

Documented outputs

GPT-6 Luna

text

OpenAI GPT-6 Luna model documentation

Qwen3.8-27B

Not sourced

Provider availability

GPT-6 Luna

Generally Available · OpenAI Responses API, OpenAI Chat Completions API, ChatGPT Work, Codex

OpenAI GPT-6 Sol and Luna launch

Qwen3.8-27B

Not sourced

Reasoning profile

GPT-6 Luna

Reasoning

Qwen3.8-27B

Reasoning

Weight access

GPT-6 Luna

Proprietary

Qwen3.8-27B

Open Weight

License

GPT-6 Luna

Proprietary

Qwen3.8-27B

Open Weight

Release date

GPT-6 Luna

2026-09-22

Qwen3.8-27B

2026-08-05

If you already use one of these models

Deployment change: The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
Quality signal: GPT-6 Luna has the higher public score estimate, 62.59 versus 54.62, but the 90% score intervals overlap.
Workload cost: A complete comparable API-rate estimate is not available for both models.
Context tradeoff: GPT-6 Luna has the larger documented window (1.05M).

Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, GPT-6 Luna or Qwen3.8-27B?

GPT-6 Luna has the higher public score estimate, 62.59 versus 54.62, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, GPT-6 Luna or Qwen3.8-27B?

GPT-6 Luna has the higher public coding point estimate, 51.8 to 46.1, with Supported evidence for both models. The conditional ranges do not establish rank confidence.

Which is better for agentic tasks, GPT-6 Luna or Qwen3.8-27B?

Qwen3.8-27B scores higher for agentic tasks on the public lane, 62.6 to 53.5. GPT-6 Luna is scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which costs less, GPT-6 Luna or Qwen3.8-27B?

Both models do not have comparable published API token rates, so this page does not name a universal price winner.

Which has the larger context window, GPT-6 Luna or Qwen3.8-27B?

GPT-6 Luna has the larger documented context window: 1.05M, compared with 262K.

Self-host vs API cost

Estimates at 50,000 req/day · 1000 tokens/req average.

GPT-6 Luna

API / mo$450

Self-host / moNot listed

Break-even—

Proprietary model — self-hosting not applicable.

Qwen3.8-27B

API / mo$0

Self-host / mo$298

Break-even—

Model the full break-even

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence55 rows

Agentic

ExploitGym
GPT-6 Luna11.6%
Source
Qwen3.8-27B—
Not directly comparable
Hermes Index
GPT-6 Luna33.9%
Source
Qwen3.8-27B—
Not directly comparable
Hermes Bench
GPT-6 Luna70.5%
Source
Qwen3.8-27B—
Not directly comparable
MazeBench
GPT-6 Luna0%
Source
Qwen3.8-27B0%
Source
Tie
OSWorld 2.1 Offline
GPT-6 Luna48.9%
Source
Qwen3.8-27B—
Not directly comparable
Terminal-Bench 2.1
GPT-6 Luna—
Qwen3.8-27B73.0%
Source
Not directly comparable
CoWorkBench
GPT-6 Luna—
Qwen3.8-27B70.7%
Source
Not directly comparable
JobBench
GPT-6 Luna—
Qwen3.8-27B33.4%
Source
Not directly comparable
Agents' Last Exam
GPT-6 Luna—
Qwen3.8-27B42.9%
Source
Not directly comparable
OSWorld-Verified
GPT-6 Luna—
Qwen3.8-27B84.3%
Source
Not directly comparable
WebArena-Verified
GPT-6 Luna—
Qwen3.8-27B64.8%
Source
Not directly comparable
AndroidWorld
GPT-6 Luna—
Qwen3.8-27B81.9%
Source
Not directly comparable
Terminal-Bench 2.1 (Vals)
GPT-6 Luna—
Qwen3.8-27B58.4%
Source
Not directly comparable

Coding

DeepSWE
GPT-6 Luna66.6%
Source
Qwen3.8-27B42.2%
Source
GPT-6 Luna leads this result
Bug Hunt Bench
Shared source
GPT-6 Luna18.3 fixes
Qwen3.8-27B15.0 fixes
GPT-6 Luna leads this result
FrontierCode 2.0
GPT-6 Luna21.9%
Source
Qwen3.8-27B—
Not directly comparable
Terminal-Bench 2.1
GPT-6 Luna—
Qwen3.8-27B73.0%
Source
Not directly comparable
SWE-bench Pro
GPT-6 Luna—
Qwen3.8-27B61.7%
Source
Not directly comparable
NL2Repo
GPT-6 Luna—
Qwen3.8-27B42.3%
Source
Not directly comparable
LiveCodeBench v6
GPT-6 Luna—
Qwen3.8-27B90.3%
Source
Not directly comparable
VulcanBench v3
GPT-6 Luna—
Qwen3.8-27B82.6%
Source
Not directly comparable
LiveCodeBench (Vals)
GPT-6 Luna—
Qwen3.8-27B84.0%
Source
Not directly comparable
SWE-bench (Vals)
GPT-6 Luna—
Qwen3.8-27B86.0%
Source
Not directly comparable

Reasoning

BioBasicsBench
GPT-6 Luna37.6%
Source
Qwen3.8-27B—
Not directly comparable
ARC-AGI-1
GPT-6 Luna86.67%
Source
Qwen3.8-27B—
Not directly comparable
ARC-AGI-2
GPT-6 Luna59.3%
Source
Qwen3.8-27B—
Not directly comparable
ARC-AGI-3
GPT-6 Luna0.1%
Source
Qwen3.8-27B—
Not directly comparable
ARC-AGI-3 Provider Adapter
GPT-6 Luna0.59%
Source
Qwen3.8-27B—
Not directly comparable

Multimodal

ExtractBench Pro
GPT-6 Luna0%
Source
Qwen3.8-27B—
Not directly comparable
ExtractBench (LlamaIndex)
Shared source
GPT-6 Luna91.01%
Qwen3.8-27B89.75%
GPT-6 Luna leads this result
ParseBench
Shared source
GPT-6 Luna64.13%
Qwen3.8-27B69.03%
Qwen3.8-27B leads this result
MathVision
GPT-6 Luna—
Qwen3.8-27B90.0%
Source
Not directly comparable
MathVision w/ Python
GPT-6 Luna—
Qwen3.8-27B94.6%
Source
Not directly comparable
BabyVision
GPT-6 Luna—
Qwen3.8-27B65.7%
Source
Not directly comparable
BabyVision w/ Python
GPT-6 Luna—
Qwen3.8-27B85.6%
Source
Not directly comparable
Vision2Web
GPT-6 Luna—
Qwen3.8-27B62.9%
Source
Not directly comparable
CharXiv w/o tools
GPT-6 Luna—
Qwen3.8-27B83.7%
Source
Not directly comparable
CharXiv
GPT-6 Luna—
Qwen3.8-27B90.2%
Source
Not directly comparable
OmniDocBench 1.5
GPT-6 Luna—
Qwen3.8-27B91.1%
Source
Not directly comparable
RealWorldQA
GPT-6 Luna—
Qwen3.8-27B85.9%
Source
Not directly comparable
ERQA
GPT-6 Luna—
Qwen3.8-27B65.5%
Source
Not directly comparable

Knowledge

HealthBench (raw)
GPT-6 Luna50.0%
Source
Qwen3.8-27B—
Not directly comparable
HealthBench (length-adjusted)
GPT-6 Luna54.5%
Source
Qwen3.8-27B—
Not directly comparable
HealthBench Professional
GPT-6 Luna60.8%
Source
Qwen3.8-27B—
Not directly comparable
HealthBench Professional (raw)
GPT-6 Luna61.2%
Source
Qwen3.8-27B—
Not directly comparable
HealthBench Hard
GPT-6 Luna31.4%
Source
Qwen3.8-27B—
Not directly comparable
GPQA
GPT-6 Luna—
Qwen3.8-27B89.2%
Source
Not directly comparable
GPQA-D
GPT-6 Luna—
Qwen3.8-27B89.2%
Source
Not directly comparable
HLE
GPT-6 Luna—
Qwen3.8-27B30.8%
Source
Not directly comparable
HLE w/o tools
GPT-6 Luna—
Qwen3.8-27B30.8%
Source
Not directly comparable
GPQA Diamond (Vals)
GPT-6 Luna—
Qwen3.8-27B88.9%
Source
Not directly comparable
MMLU-Pro (Vals)
GPT-6 Luna—
Qwen3.8-27B84.3%
Source
Not directly comparable

Instruction following

IFBench
GPT-6 Luna—
Qwen3.8-27B79.5%
Source
Not directly comparable

Math

FrontierMath v2 (Tiers 1-3)
GPT-6 Luna78.947%
Source
Qwen3.8-27B—
Not directly comparable
FrontierMath v2 (Tier 4)
GPT-6 Luna56.098%
Source
Qwen3.8-27B—
Not directly comparable

55 public results · 5 shared

Related comparisons

Compare API pricing Read the methodology Open the model selector

Watch GPT-6 Luna vs Qwen3.8-27B

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 6,100+ subscribers.

Last updated November 6, 2026

GPT-6 Luna vs Qwen3.8-27B

Decision reading

Which one for your work

Coding work

Long documents

Agentic work

Chat turn cost

Cache-heavy agent loop cost

Repository review cost

Which one for a specific job

What is actually comparable

Shape of the matched evidence

Category results, on a stated basis

Coding

Knowledge

Agentic

Reasoning

Multimodal

Multilingual

Instruction following

Math

What each workload costs

Chat turn

Repository review

Cache-heavy agent loop

Specification differences

Context window

API model ID

Cached-input rate

Documented inputs

Documented outputs

Provider availability

Reasoning profile

Weight access

License

Release date

If you already use one of these models

Questions

Which is better, GPT-6 Luna or Qwen3.8-27B?

Which is better for coding, GPT-6 Luna or Qwen3.8-27B?

Which is better for agentic tasks, GPT-6 Luna or Qwen3.8-27B?

Which costs less, GPT-6 Luna or Qwen3.8-27B?

Which has the larger context window, GPT-6 Luna or Qwen3.8-27B?

Self-host vs API cost

Benchmark evidence

Agentic

Coding

Reasoning

Multimodal

Knowledge

Instruction following

Math

Related comparisons

Watch GPT-6 Luna vs Qwen3.8-27B