ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

Qwen3 235B-A22BvsQwen3.8 2.4T-A95B

Qwen3.8 2.4T-A95B leads 1 of 1 shared benchmark. Qwen3.8 2.4T-A95B has the larger context window (1M tokens).

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
0 – 1
Qwen3.8 2.4T-A95B leads
Cheaper per token
list price, input + output
Larger context
Qwen3.8 2.4T-A95B
1M tokens
Provider uptime (30d)
Qwen
Qwen3 235B-A22B
Qwen · released 2025-04-29
text
Context
262K
Max out
Input /1M
Output /1M
Cached /1M
Scores
4 · 3 core
Qwen3.8 2.4T-A95B
Qwen · released 2026-08-12
text
Context
1M
Max out
131K
Input /1M
$1.65
Output /1M
$4.95
Cached /1M
Scores
4 · 1 core
Quality

Benchmark matrix

BenchmarkQwen3 235B-A22BQwen3.8 2.4T-A95BΔEdge
Reported by both · 1
GPQA Diamond70.7% epoch92.6% -21.9 ptQwen3.8 2.4T-A95B
Only Qwen3 235B-A22B reports · 3
AIME 202580.8% matharenanot reported
LMArena Elo1366.1 not reported
ARC-AGI-1 (pass@1)40.75% not reported
Only Qwen3.8 2.4T-A95B reports · 3
BabyVisionnot reported82%
OSWorldnot reported86.1%
PaperBenchnot reported93%
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Qwen3 235B-A22B minus Qwen3.8 2.4T-A95B in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing

Side by side

SpecQwen3 235B-A22BQwen3.8 2.4T-A95BEdge
Context window262K tokens1M tokensQwen3.8 2.4T-A95B
Max output131K tokens
Input price / 1M$1.65
Output price / 1M$4.95
Cached input / 1M
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
$6.6
ModalitiestexttextTie
Released2025-04-292026-08-12
Cited benchmark scores44
Reliability

Qwen status

All providers
More matchups

Qwen3 235B-A22B vs …

More matchups

Qwen3.8 2.4T-A95B vs …

Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run Qwen3 235B-A22B and Qwen3.8 2.4T-A95B on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.