Coding · Benchmark profile

SWE-bench Verified

A human-validated subset of real GitHub software-engineering issue resolution tasks.

Ranking weightedVerifiedActive21 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on SWE-bench Verified

Leaderboard · 21 models

Claude Mythos 5 leads at 95.5%, followed by Claude Fable 5 (95%) and Claude Opus 4.8 (88.6%).

Ranking · 3.2%CodingVerified

#1Closed

Claude Mythos 5

Anthropic

claude-mythos-5

95.5%

Overall #1 · Context 1M tokens

#2Closed

Claude Fable 5

Anthropic

claude-fable-5

95%

Overall #2 · Context 1M tokens

#3Closed

Claude Opus 4.8

Anthropic

claude-opus-4-8

88.6%

Overall #5 · Context 1M tokens

Visual analysis

See the field, not just the table.

Three views of the same sourced leaderboard: model placement, the score curve, and descriptive provider averages. These visuals never alter ranking calculations.

ModelsColours

AnthropicOpenAIMiniMaxAlibaba CloudTencentXiaomiZ.AIMetaMoonshot AIDeepSeek

Leaderboard bars

20 ranked models · higher is better

View accessible chart data

Accessible data for Leaderboard bars
Model	Rank	Provider	Score
Claude Mythos 5	#1	Anthropic	95.5%
Claude Fable 5	#2	Anthropic	95%
Claude Opus 4.8	#3	Anthropic	88.6%
Claude Sonnet 5	#4	Anthropic	85.2%
GPT-5.3-Codex	#5	OpenAI	85%
Claude Opus 4.5	#6	Anthropic	80.9%
Claude Opus 4.6	#7	Anthropic	80.8%
MiniMax M3	#8	MiniMax	80.5%
Qwen3.7-Max	#9	Alibaba Cloud	80.4%
Claude Sonnet 4.6	#10	Anthropic	79.6%
Qwen3.6 Plus	#11	Alibaba Cloud	78.8%
Hy3	#12	Tencent	78%
MiMo-V2-Pro	#13	Xiaomi	78%
GLM-5	#14	Z.AI	77.8%
Qwen3.7-Plus	#15	Alibaba Cloud	77.7%
Muse Spark	#16	Meta	77.4%
Kimi K2.5	#17	Moonshot AI	76.8%
MiMo-V2-Omni	#18	Xiaomi	74.8%
DeepSeek V4 Flash	#19	DeepSeek	73.7%
DeepSeek V4 Pro	#20	DeepSeek	73.6%

Score curve

Shows how quickly scores separate down the field

View accessible chart data

Accessible data for Score curve
Model	Rank	Provider	Score
Claude Mythos 5	#1	Anthropic	95.5%
Claude Fable 5	#2	Anthropic	95%
Claude Opus 4.8	#3	Anthropic	88.6%
Claude Sonnet 5	#4	Anthropic	85.2%
GPT-5.3-Codex	#5	OpenAI	85%
Claude Opus 4.5	#6	Anthropic	80.9%
Claude Opus 4.6	#7	Anthropic	80.8%
MiniMax M3	#8	MiniMax	80.5%
Qwen3.7-Max	#9	Alibaba Cloud	80.4%
Claude Sonnet 4.6	#10	Anthropic	79.6%
Qwen3.6 Plus	#11	Alibaba Cloud	78.8%
Hy3	#12	Tencent	78%
MiMo-V2-Pro	#13	Xiaomi	78%
GLM-5	#14	Z.AI	77.8%
Qwen3.7-Plus	#15	Alibaba Cloud	77.7%
Muse Spark	#16	Meta	77.4%
Kimi K2.5	#17	Moonshot AI	76.8%
MiMo-V2-Omni	#18	Xiaomi	74.8%
DeepSeek V4 Flash	#19	DeepSeek	73.7%
DeepSeek V4 Pro	#20	DeepSeek	73.6%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Anthropic	7	86.5%
OpenAI	1	85%
MiniMax	1	80.5%
Alibaba Cloud	3	79.0%
Tencent	1	78%
Z.AI	1	77.8%
Meta	1	77.4%
Moonshot AI	1	76.8%
Xiaomi	2	76.4%
DeepSeek	2	73.7%

Benchmark score table (21 models)

One best score per model · higher is better

SWE-bench Verified leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	Claude Mythos 5 claude-mythos-5	Anthropic	closed	95.5%
#2	Claude Fable 5 claude-fable-5	Anthropic	closed	95%
#3	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	88.6%
#4	Claude Sonnet 5 claude-sonnet-5	Anthropic	closed	85.2%
#5	GPT-5.3-Codex gpt-5.3-codex	OpenAI	closed	85%
#6	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	80.9%
#7	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	80.8%
#8	MiniMax M3 MiniMax-M3	MiniMax	—	80.5%
#9	Qwen3.7-Max qwen3.7-max	Alibaba Cloud	—	80.4%
#10	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	79.6%
#11	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	78.8%
#12	Hy3 hy3	Tencent	open	78%
#13	MiMo-V2-Pro mimo-v2-pro	Xiaomi	closed	78%
#14	GLM-5 glm-5	Z.AI	open	77.8%
#15	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	77.7%
#16	Muse Spark muse-spark	Meta	closed	77.4%
#17	Kimi K2.5 kimi-k2.5	Moonshot AI	—	76.8%
#18	MiMo-V2-Omni mimo-v2-omni	Xiaomi	closed	74.8%
#19	DeepSeek V4 Flash deepseek-v4-flash	DeepSeek	open	73.7%
#20	DeepSeek V4 Pro deepseek-v4-pro	DeepSeek	open	73.6%
#21	Claude Haiku 4.5 claude-haiku-4-5-20251001	Anthropic	closed	73.3%

The top of this snapshot is led by Claude Mythos 5 at 95.5%; third place is 6.9 points behind. The top-10 spread is 15.9 points.

Compare top models

Claude Mythos 5 vs Claude Fable 5 Claude Fable 5 vs Claude Opus 4.8 Claude Opus 4.8 vs Claude Sonnet 5 Claude Sonnet 5 vs GPT-5.3-Codex

About SWE-bench Verified

Definition and scoring

Organisation: SWE-bench authors
Category: Coding
Version: Verified
Direction: higher is better
Ranking use: Weighted 3.2%
Contamination risk: Medium

A human-validated subset of real GitHub software-engineering issue resolution tasks. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.

Open benchmark source ↗

FAQ

Quick answers

What does SWE-bench Verified measure?

A human-validated subset of real GitHub software-engineering issue resolution tasks.

Which model scores highest on SWE-bench Verified?

Claude Mythos 5 by Anthropic currently leads with 95.5%.

How many models are evaluated on SWE-bench Verified?

21 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 3.2% of its capability category. Claude Mythos 5 is overall #1.

LiveCodeBench

Rolling · 38 results · Weighted 3.2%

CRUXEval

Rolling · Not evaluated · Not Weighted

SciCode

Rolling · 92 results · Weighted 3.2%

BigCodeBench

Rolling · Not evaluated · Not Weighted

SWE-Bench Pro

v1 · 33 results · Weighted 3.2%

CursorBench

3.2 · 20 results · Weighted 3.2%

Coding · Benchmark profile

SWE-bench Verified

A human-validated subset of real GitHub software-engineering issue resolution tasks.

Ranking weightedVerifiedActive21 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on SWE-bench Verified

Leaderboard · 21 models

Claude Mythos 5 leads at 95.5%, followed by Claude Fable 5 (95%) and Claude Opus 4.8 (88.6%).

Ranking · 3.2%CodingVerified

#1Closed

Claude Mythos 5

Anthropic

claude-mythos-5

95.5%

Overall #1 · Context 1M tokens

#2Closed

Claude Fable 5

Anthropic

claude-fable-5

95%

Overall #2 · Context 1M tokens

#3Closed

Claude Opus 4.8

Anthropic

claude-opus-4-8

88.6%

Overall #5 · Context 1M tokens

Visual analysis

See the field, not just the table.

Three views of the same sourced leaderboard: model placement, the score curve, and descriptive provider averages. These visuals never alter ranking calculations.

ModelsColours

AnthropicOpenAIMiniMaxAlibaba CloudTencentXiaomiZ.AIMetaMoonshot AIDeepSeek

Leaderboard bars

20 ranked models · higher is better

View accessible chart data

Accessible data for Leaderboard bars
Model	Rank	Provider	Score
Claude Mythos 5	#1	Anthropic	95.5%
Claude Fable 5	#2	Anthropic	95%
Claude Opus 4.8	#3	Anthropic	88.6%
Claude Sonnet 5	#4	Anthropic	85.2%
GPT-5.3-Codex	#5	OpenAI	85%
Claude Opus 4.5	#6	Anthropic	80.9%
Claude Opus 4.6	#7	Anthropic	80.8%
MiniMax M3	#8	MiniMax	80.5%
Qwen3.7-Max	#9	Alibaba Cloud	80.4%
Claude Sonnet 4.6	#10	Anthropic	79.6%
Qwen3.6 Plus	#11	Alibaba Cloud	78.8%
Hy3	#12	Tencent	78%
MiMo-V2-Pro	#13	Xiaomi	78%
GLM-5	#14	Z.AI	77.8%
Qwen3.7-Plus	#15	Alibaba Cloud	77.7%
Muse Spark	#16	Meta	77.4%
Kimi K2.5	#17	Moonshot AI	76.8%
MiMo-V2-Omni	#18	Xiaomi	74.8%
DeepSeek V4 Flash	#19	DeepSeek	73.7%
DeepSeek V4 Pro	#20	DeepSeek	73.6%

Score curve

Shows how quickly scores separate down the field

View accessible chart data

Accessible data for Score curve
Model	Rank	Provider	Score
Claude Mythos 5	#1	Anthropic	95.5%
Claude Fable 5	#2	Anthropic	95%
Claude Opus 4.8	#3	Anthropic	88.6%
Claude Sonnet 5	#4	Anthropic	85.2%
GPT-5.3-Codex	#5	OpenAI	85%
Claude Opus 4.5	#6	Anthropic	80.9%
Claude Opus 4.6	#7	Anthropic	80.8%
MiniMax M3	#8	MiniMax	80.5%
Qwen3.7-Max	#9	Alibaba Cloud	80.4%
Claude Sonnet 4.6	#10	Anthropic	79.6%
Qwen3.6 Plus	#11	Alibaba Cloud	78.8%
Hy3	#12	Tencent	78%
MiMo-V2-Pro	#13	Xiaomi	78%
GLM-5	#14	Z.AI	77.8%
Qwen3.7-Plus	#15	Alibaba Cloud	77.7%
Muse Spark	#16	Meta	77.4%
Kimi K2.5	#17	Moonshot AI	76.8%
MiMo-V2-Omni	#18	Xiaomi	74.8%
DeepSeek V4 Flash	#19	DeepSeek	73.7%
DeepSeek V4 Pro	#20	DeepSeek	73.6%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Anthropic	7	86.5%
OpenAI	1	85%
MiniMax	1	80.5%
Alibaba Cloud	3	79.0%
Tencent	1	78%
Z.AI	1	77.8%
Meta	1	77.4%
Moonshot AI	1	76.8%
Xiaomi	2	76.4%
DeepSeek	2	73.7%

Benchmark score table (21 models)

One best score per model · higher is better

SWE-bench Verified leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	Claude Mythos 5 claude-mythos-5	Anthropic	closed	95.5%
#2	Claude Fable 5 claude-fable-5	Anthropic	closed	95%
#3	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	88.6%
#4	Claude Sonnet 5 claude-sonnet-5	Anthropic	closed	85.2%
#5	GPT-5.3-Codex gpt-5.3-codex	OpenAI	closed	85%
#6	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	80.9%
#7	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	80.8%
#8	MiniMax M3 MiniMax-M3	MiniMax	—	80.5%
#9	Qwen3.7-Max qwen3.7-max	Alibaba Cloud	—	80.4%
#10	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	79.6%
#11	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	78.8%
#12	Hy3 hy3	Tencent	open	78%
#13	MiMo-V2-Pro mimo-v2-pro	Xiaomi	closed	78%
#14	GLM-5 glm-5	Z.AI	open	77.8%
#15	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	77.7%
#16	Muse Spark muse-spark	Meta	closed	77.4%
#17	Kimi K2.5 kimi-k2.5	Moonshot AI	—	76.8%
#18	MiMo-V2-Omni mimo-v2-omni	Xiaomi	closed	74.8%
#19	DeepSeek V4 Flash deepseek-v4-flash	DeepSeek	open	73.7%
#20	DeepSeek V4 Pro deepseek-v4-pro	DeepSeek	open	73.6%
#21	Claude Haiku 4.5 claude-haiku-4-5-20251001	Anthropic	closed	73.3%

The top of this snapshot is led by Claude Mythos 5 at 95.5%; third place is 6.9 points behind. The top-10 spread is 15.9 points.

Compare top models

Claude Mythos 5 vs Claude Fable 5 Claude Fable 5 vs Claude Opus 4.8 Claude Opus 4.8 vs Claude Sonnet 5 Claude Sonnet 5 vs GPT-5.3-Codex

About SWE-bench Verified

Definition and scoring

Organisation: SWE-bench authors
Category: Coding
Version: Verified
Direction: higher is better
Ranking use: Weighted 3.2%
Contamination risk: Medium

Open benchmark source ↗

FAQ

Quick answers

What does SWE-bench Verified measure?

A human-validated subset of real GitHub software-engineering issue resolution tasks.

Which model scores highest on SWE-bench Verified?

Claude Mythos 5 by Anthropic currently leads with 95.5%.

How many models are evaluated on SWE-bench Verified?

21 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 3.2% of its capability category. Claude Mythos 5 is overall #1.

LiveCodeBench

Rolling · 38 results · Weighted 3.2%

CRUXEval

Rolling · Not evaluated · Not Weighted

SciCode

Rolling · 92 results · Weighted 3.2%

BigCodeBench

Rolling · Not evaluated · Not Weighted

SWE-Bench Pro

v1 · 33 results · Weighted 3.2%

CursorBench

3.2 · 20 results · Weighted 3.2%

Leaderboard · 21 models

Claude Mythos 5

Claude Fable 5

Claude Opus 4.8

See the field, not just the table.

Benchmark score table (21 models)

Compare top models

Definition and scoring

Quick answers

What does SWE-bench Verified measure?

Which model scores highest on SWE-bench Verified?

How many models are evaluated on SWE-bench Verified?

Does this affect overall Lumina rank?

More Coding benchmarks

Leaderboard · 21 models

Claude Mythos 5

Claude Fable 5

Claude Opus 4.8

See the field, not just the table.

Benchmark score table (21 models)

Compare top models

Definition and scoring

Quick answers

What does SWE-bench Verified measure?

Which model scores highest on SWE-bench Verified?

How many models are evaluated on SWE-bench Verified?

Does this affect overall Lumina rank?

More Coding benchmarks