Coding · Benchmark profile

BenchLM Coding prior

Category prior from the BenchLM overall leaderboard coding score used to estimate missing coding coverage.

Ranking weightedbench-align-v5.1Active34 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on BenchLM Coding prior

Leaderboard · 34 models

Claude Mythos 5 leads at 82.0%, followed by Claude Fable 5 (81.7%) and GPT-5.6 Sol (74.1%).

Ranking · 1.4%Codingbench-align-v5.1

#1Closed

Claude Mythos 5

Anthropic

claude-mythos-5

82.0%

Overall #1 · Context 1M tokens

#2Closed

Claude Fable 5

Anthropic

claude-fable-5

81.7%

Overall #2 · Context 1M tokens

#3Closed

GPT-5.6 Sol

OpenAI

gpt-5.6-sol

74.1%

Overall #4 · Context 1.05M 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

AnthropicOpenAIxAIMetaGoogleXiaomiZ.AIAlibaba Cloud

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	82.0%
Claude Fable 5	#2	Anthropic	81.7%
GPT-5.6 Sol	#3	OpenAI	74.1%
Claude Opus 4.8	#4	Anthropic	73.5%
GPT-5.5	#5	OpenAI	71.4%
Grok 4.5	#6	xAI	71.3%
Claude Opus 4.7	#7	Anthropic	69.4%
Muse Spark 1.1	#8	Meta	69.1%
GPT-5.6 Terra	#9	OpenAI	66.6%
Muse Spark	#10	Meta	66.4%
Claude Opus 4.6	#11	Anthropic	65.7%
Gemini 3.5 Flash	#12	Google	65.4%
Claude Sonnet 5	#13	Anthropic	64.7%
GPT-5.6 Luna	#14	OpenAI	64.3%
GPT-5.4	#15	OpenAI	64.2%
GPT-5.3-Codex	#16	OpenAI	64.1%
Gemini 3 Pro	#17	Google	63.9%
MiMo-V2-Pro	#18	Xiaomi	63.6%
GLM-5.2	#19	Z.AI	62.3%
Qwen3.6 Plus	#20	Alibaba Cloud	60.5%

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	82.0%
Claude Fable 5	#2	Anthropic	81.7%
GPT-5.6 Sol	#3	OpenAI	74.1%
Claude Opus 4.8	#4	Anthropic	73.5%
GPT-5.5	#5	OpenAI	71.4%
Grok 4.5	#6	xAI	71.3%
Claude Opus 4.7	#7	Anthropic	69.4%
Muse Spark 1.1	#8	Meta	69.1%
GPT-5.6 Terra	#9	OpenAI	66.6%
Muse Spark	#10	Meta	66.4%
Claude Opus 4.6	#11	Anthropic	65.7%
Gemini 3.5 Flash	#12	Google	65.4%
Claude Sonnet 5	#13	Anthropic	64.7%
GPT-5.6 Luna	#14	OpenAI	64.3%
GPT-5.4	#15	OpenAI	64.2%
GPT-5.3-Codex	#16	OpenAI	64.1%
Gemini 3 Pro	#17	Google	63.9%
MiMo-V2-Pro	#18	Xiaomi	63.6%
GLM-5.2	#19	Z.AI	62.3%
Qwen3.6 Plus	#20	Alibaba Cloud	60.5%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Anthropic	6	72.8%
xAI	1	71.3%
Meta	2	67.7%
OpenAI	6	67.4%
Google	2	64.6%
Xiaomi	1	63.6%
Z.AI	1	62.3%
Alibaba Cloud	1	60.5%

Benchmark score table (34 models)

One best score per model · higher is better

BenchLM Coding prior leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	Claude Mythos 5 claude-mythos-5	Anthropic	closed	82.0%
#2	Claude Fable 5 claude-fable-5	Anthropic	closed	81.7%
#3	GPT-5.6 Sol gpt-5.6-sol	OpenAI	closed	74.1%
#4	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	73.5%
#5	GPT-5.5 gpt-5.5	OpenAI	closed	71.4%
#6	Grok 4.5 grok-4.5	xAI	closed	71.3%
#7	Claude Opus 4.7 claude-opus-4-7	Anthropic	closed	69.4%
#8	Muse Spark 1.1 muse-spark-1-1	Meta	closed	69.1%
#9	GPT-5.6 Terra gpt-5.6-terra	OpenAI	closed	66.6%
#10	Muse Spark muse-spark	Meta	closed	66.4%
#11	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	65.7%
#12	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	65.4%
#13	Claude Sonnet 5 claude-sonnet-5	Anthropic	closed	64.7%
#14	GPT-5.6 Luna gpt-5.6-luna	OpenAI	closed	64.3%
#15	GPT-5.4 gpt-5.4	OpenAI	closed	64.2%
#16	GPT-5.3-Codex gpt-5.3-codex	OpenAI	closed	64.1%
#17	Gemini 3 Pro gemini-3-pro	Google	closed	63.9%
#18	MiMo-V2-Pro mimo-v2-pro	Xiaomi	closed	63.6%
#19	GLM-5.2 glm-5-2	Z.AI	open	62.3%
#20	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	60.5%
#21	GLM-5 glm-5	Z.AI	open	60.3%
#22	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	60.0%
#23	GLM-5V-Turbo glm-5v-turbo	Z.AI	closed	59.9%
#24	DeepSeek V4 Pro deepseek-v4-pro	DeepSeek	open	59.3%
#25	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	58.7%
#26	MiniMax M3 MiniMax-M3	MiniMax	—	58.1%
#27	GLM-5.1 glm-5-1	Z.AI	open	56.7%
#28	MiMo-V2.5-Pro mimo-v2-5-pro	Xiaomi	closed	55.2%
#29	GPT-5.4 nano gpt.5.4.nano	OpenAI	closed	51.6%
#30	GPT-5.1 gpt.5.1	OpenAI	closed	51.0%
#31	Gemma 4 31B gemma-4-31b	Google	open	50.1%
#32	Gemini 3 Flash gemini-3-flash	Google	closed	45.9%
#33	MiniMax M2.7 minimax-m2-7	MiniMax	open	44.7%
#34	Grok 4.3 grok-4.3	xAI	closed	40.2%

The top of this snapshot is led by Claude Mythos 5 at 82.0%; third place is 7.9 points behind. The top-10 spread is 15.6 points.

Compare top models

Claude Mythos 5 vs Claude Fable 5 Claude Fable 5 vs GPT-5.6 Sol GPT-5.6 Sol vs Claude Opus 4.8 Claude Opus 4.8 vs GPT-5.5

About BenchLM Coding prior

Definition and scoring

Organisation: BenchLM
Category: Coding
Version: bench-align-v5.1
Direction: higher is better
Ranking use: Weighted 1.4%
Contamination risk: Unknown

Category prior from the BenchLM overall leaderboard coding score used to estimate missing coding coverage. 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 BenchLM Coding prior measure?

Category prior from the BenchLM overall leaderboard coding score used to estimate missing coding coverage.

Which model scores highest on BenchLM Coding prior?

Claude Mythos 5 by Anthropic currently leads with 82.0%.

How many models are evaluated on BenchLM Coding prior?

34 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 1.4% of its capability category. Claude Mythos 5 is overall #1.

LiveCodeBench

Rolling · 38 results · Weighted 3.2%

SWE-bench Verified

Verified · 21 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%

Coding · Benchmark profile

BenchLM Coding prior

Category prior from the BenchLM overall leaderboard coding score used to estimate missing coding coverage.

Ranking weightedbench-align-v5.1Active34 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on BenchLM Coding prior

Leaderboard · 34 models

Claude Mythos 5 leads at 82.0%, followed by Claude Fable 5 (81.7%) and GPT-5.6 Sol (74.1%).

Ranking · 1.4%Codingbench-align-v5.1

#1Closed

Claude Mythos 5

Anthropic

claude-mythos-5

82.0%

Overall #1 · Context 1M tokens

#2Closed

Claude Fable 5

Anthropic

claude-fable-5

81.7%

Overall #2 · Context 1M tokens

#3Closed

GPT-5.6 Sol

OpenAI

gpt-5.6-sol

74.1%

Overall #4 · Context 1.05M 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

AnthropicOpenAIxAIMetaGoogleXiaomiZ.AIAlibaba Cloud

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	82.0%
Claude Fable 5	#2	Anthropic	81.7%
GPT-5.6 Sol	#3	OpenAI	74.1%
Claude Opus 4.8	#4	Anthropic	73.5%
GPT-5.5	#5	OpenAI	71.4%
Grok 4.5	#6	xAI	71.3%
Claude Opus 4.7	#7	Anthropic	69.4%
Muse Spark 1.1	#8	Meta	69.1%
GPT-5.6 Terra	#9	OpenAI	66.6%
Muse Spark	#10	Meta	66.4%
Claude Opus 4.6	#11	Anthropic	65.7%
Gemini 3.5 Flash	#12	Google	65.4%
Claude Sonnet 5	#13	Anthropic	64.7%
GPT-5.6 Luna	#14	OpenAI	64.3%
GPT-5.4	#15	OpenAI	64.2%
GPT-5.3-Codex	#16	OpenAI	64.1%
Gemini 3 Pro	#17	Google	63.9%
MiMo-V2-Pro	#18	Xiaomi	63.6%
GLM-5.2	#19	Z.AI	62.3%
Qwen3.6 Plus	#20	Alibaba Cloud	60.5%

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	82.0%
Claude Fable 5	#2	Anthropic	81.7%
GPT-5.6 Sol	#3	OpenAI	74.1%
Claude Opus 4.8	#4	Anthropic	73.5%
GPT-5.5	#5	OpenAI	71.4%
Grok 4.5	#6	xAI	71.3%
Claude Opus 4.7	#7	Anthropic	69.4%
Muse Spark 1.1	#8	Meta	69.1%
GPT-5.6 Terra	#9	OpenAI	66.6%
Muse Spark	#10	Meta	66.4%
Claude Opus 4.6	#11	Anthropic	65.7%
Gemini 3.5 Flash	#12	Google	65.4%
Claude Sonnet 5	#13	Anthropic	64.7%
GPT-5.6 Luna	#14	OpenAI	64.3%
GPT-5.4	#15	OpenAI	64.2%
GPT-5.3-Codex	#16	OpenAI	64.1%
Gemini 3 Pro	#17	Google	63.9%
MiMo-V2-Pro	#18	Xiaomi	63.6%
GLM-5.2	#19	Z.AI	62.3%
Qwen3.6 Plus	#20	Alibaba Cloud	60.5%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Anthropic	6	72.8%
xAI	1	71.3%
Meta	2	67.7%
OpenAI	6	67.4%
Google	2	64.6%
Xiaomi	1	63.6%
Z.AI	1	62.3%
Alibaba Cloud	1	60.5%

Benchmark score table (34 models)

One best score per model · higher is better

BenchLM Coding prior leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	Claude Mythos 5 claude-mythos-5	Anthropic	closed	82.0%
#2	Claude Fable 5 claude-fable-5	Anthropic	closed	81.7%
#3	GPT-5.6 Sol gpt-5.6-sol	OpenAI	closed	74.1%
#4	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	73.5%
#5	GPT-5.5 gpt-5.5	OpenAI	closed	71.4%
#6	Grok 4.5 grok-4.5	xAI	closed	71.3%
#7	Claude Opus 4.7 claude-opus-4-7	Anthropic	closed	69.4%
#8	Muse Spark 1.1 muse-spark-1-1	Meta	closed	69.1%
#9	GPT-5.6 Terra gpt-5.6-terra	OpenAI	closed	66.6%
#10	Muse Spark muse-spark	Meta	closed	66.4%
#11	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	65.7%
#12	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	65.4%
#13	Claude Sonnet 5 claude-sonnet-5	Anthropic	closed	64.7%
#14	GPT-5.6 Luna gpt-5.6-luna	OpenAI	closed	64.3%
#15	GPT-5.4 gpt-5.4	OpenAI	closed	64.2%
#16	GPT-5.3-Codex gpt-5.3-codex	OpenAI	closed	64.1%
#17	Gemini 3 Pro gemini-3-pro	Google	closed	63.9%
#18	MiMo-V2-Pro mimo-v2-pro	Xiaomi	closed	63.6%
#19	GLM-5.2 glm-5-2	Z.AI	open	62.3%
#20	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	60.5%
#21	GLM-5 glm-5	Z.AI	open	60.3%
#22	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	60.0%
#23	GLM-5V-Turbo glm-5v-turbo	Z.AI	closed	59.9%
#24	DeepSeek V4 Pro deepseek-v4-pro	DeepSeek	open	59.3%
#25	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	58.7%
#26	MiniMax M3 MiniMax-M3	MiniMax	—	58.1%
#27	GLM-5.1 glm-5-1	Z.AI	open	56.7%
#28	MiMo-V2.5-Pro mimo-v2-5-pro	Xiaomi	closed	55.2%
#29	GPT-5.4 nano gpt.5.4.nano	OpenAI	closed	51.6%
#30	GPT-5.1 gpt.5.1	OpenAI	closed	51.0%
#31	Gemma 4 31B gemma-4-31b	Google	open	50.1%
#32	Gemini 3 Flash gemini-3-flash	Google	closed	45.9%
#33	MiniMax M2.7 minimax-m2-7	MiniMax	open	44.7%
#34	Grok 4.3 grok-4.3	xAI	closed	40.2%

The top of this snapshot is led by Claude Mythos 5 at 82.0%; third place is 7.9 points behind. The top-10 spread is 15.6 points.

Compare top models

Claude Mythos 5 vs Claude Fable 5 Claude Fable 5 vs GPT-5.6 Sol GPT-5.6 Sol vs Claude Opus 4.8 Claude Opus 4.8 vs GPT-5.5

About BenchLM Coding prior

Definition and scoring

Organisation: BenchLM
Category: Coding
Version: bench-align-v5.1
Direction: higher is better
Ranking use: Weighted 1.4%
Contamination risk: Unknown

Open benchmark source ↗

FAQ

Quick answers

What does BenchLM Coding prior measure?

Category prior from the BenchLM overall leaderboard coding score used to estimate missing coding coverage.

Which model scores highest on BenchLM Coding prior?

Claude Mythos 5 by Anthropic currently leads with 82.0%.

How many models are evaluated on BenchLM Coding prior?

34 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 1.4% of its capability category. Claude Mythos 5 is overall #1.

LiveCodeBench

Rolling · 38 results · Weighted 3.2%

SWE-bench Verified

Verified · 21 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%

Leaderboard · 34 models

Claude Mythos 5

Claude Fable 5

GPT-5.6 Sol

See the field, not just the table.

Benchmark score table (34 models)

Compare top models

Definition and scoring

Quick answers

What does BenchLM Coding prior measure?

Which model scores highest on BenchLM Coding prior?

How many models are evaluated on BenchLM Coding prior?

Does this affect overall Lumina rank?

More Coding benchmarks

Leaderboard · 34 models

Claude Mythos 5

Claude Fable 5

GPT-5.6 Sol

See the field, not just the table.

Benchmark score table (34 models)

Compare top models

Definition and scoring

Quick answers

What does BenchLM Coding prior measure?

Which model scores highest on BenchLM Coding prior?

How many models are evaluated on BenchLM Coding prior?

Does this affect overall Lumina rank?

More Coding benchmarks