Compare models

Price, context, and benchmarks side by side

Context window, price per million tokens, reasoning levels, and benchmark results with the date they were checked. Unconfirmed fields are marked; provider pricing and capability details can change.

This is the full comparison table for model APIs (GPT, Claude, Gemini, and so on). For launches in the last 28 days, open New models. For Cursor, Claude Code, Muse Code, Lovable and similar products, see Agentic harnesses. Right now that includes 13 recent launches.

Run the model picker4 questions → frontier, balanced, or budget shortlist

Launch guides Open weights Benchmark glossary Switch guides Popular this week Category picks Replacement maps

Task guides

Choose by the work you need done

Short guides for coding, writing, research, RAG, voice, and more — then compare the models that actually fit.

All AI guides

Task guide · coding

Best AI for coding: choose the right model and coding agent

Alternatives guide · Cursor

Cursor alternatives: compare editor, terminal, and open-source agents

Alternatives guide · open source

Open-source coding agent alternatives: choose the client, model, and privacy boundary

Task guide · writing

Best AI for writing: match quality, voice, and review effort

Task guide · research

Best AI for research: compare context, sources, and verification

Task guide · meetings

Best AI for meeting notes: optimise for capture, privacy, and follow-through

Tier

Weights

Reasoning controls

AI model tier, context, output, pricing, and weight comparison
Model	Tier	Context	Max output	In / 1M	Out / 1M	Weights
GPT-5.6 SolOpenAI	Frontier	1.05M	128K	$5	$30	Closed
GPT-5.6 TerraOpenAI	Balanced	1.05M	128K	$2	$12	Closed
GPT-5.6 LunaOpenAI	Budget	1.05M	128K	$0.20	$1.20	Closed
Claude Opus 5Anthropic	Frontier	1M	128K	$5	$25	Closed
Claude Fable 5Anthropic	Frontier	1M	128K	$10	$50	Closed
Claude Sonnet 5Anthropic	Balanced	1M	128K	$2	$10	Closed
Gemini 3.1 ProGoogle	Frontier	1.05M	66K	$2	$12	Closed
Gemini 3.6 FlashGoogle	Budget	1.05M	66K	$0.75	$3.75	Closed
Gemini 3.7 FlashGoogle	Budget	1.05M	66K	$0.75	$3.75	Closed
Gemini 3.5 Flash-LiteGoogle	Budget	1.05M	66K	$0.30	$2.50	Closed
Kimi K3Moonshot AI	Frontier	1.05M	1M	$3	$15	Open
Kimi K2.6Moonshot AI	Budget	262K	Not verified	$0.95	$4	Open
GLM 5.2Z.ai	Frontier	1M	128K	$1.40	$4.40	Open
GLM 5.3Z.ai	Frontier	1M	128K	Not verified	Not verified	Closed
Grok 4.5SpaceXAI	Balanced	500K	Not verified	$2	$6	Closed
Grok 4.6SpaceXAI	Frontier	500K	Not verified	$2	$6	Closed
Claude Haiku 4.5Anthropic	Budget	200K	64K	$1	$5	Closed
DeepSeek V4-FlashDeepSeek	Budget	1M	384K	$0.14	$0.28	Open
DeepSeek V4-ProDeepSeek	Balanced	1M	384K	$0.435	$0.87	Open
Ling 3.0 FlashInclusionAI	Budget	262K	Not verified	$0.075	$0.22	Open
Qwen3.8-MaxQwen	Frontier	1M	131K	$2	$6	Closed
Qwen3.8-2.4T-A95BQwen	Frontier	262K	131K	Not verified	Not verified	Open
Qwen3.8-27BQwen	Balanced	262K	131K	Not verified	Not verified	Open
Muse Spark 1.2Meta	Frontier	1.05M	Not verified	$1.25	$4.25	Closed
Muse Glimmer 30BMeta	Balanced	131K	Not verified	$0	$0	Open

Compare two models

Left

Right

Open comparison

Prices are USD per million tokens at standard rates, excluding batch and caching discounts. Last verified August 14, 2026. Model pricing and capability in this category change frequently — check the provider before committing spend.

Benchmark

DeepSWE 1.1 — coding agent performance and cost

Score alone hides the decision. Cost per completed task sits alongside it, because a model a few points behind at a fraction of the price is usually the better build.

Local leader

Claude Opus 5 [max]

74%

Rows shown

Highest published reasoning effort per model (not best Pass@1)

Snapshot date

2026-08-13

Mirrored from deepswe.datacurve.ai

Better is toward the top-right (higher pass rate, lower cost). X-axis is reversed to match DeepSWE’s public chart. v1.1 uses average cost / tokens / steps; v1 uses published medians.

DeepSWE 1.1 leaderboard with pass rate, cost, tokens, and steps per task
#	Model	Pass@1	Cost / task	Tokens / task	Steps / task
1	Claude Opus 5 [max]	74%	$11.84	118k	99
2	GPT-5.6 Sol [max]	73%	$8.39	60k	61
3	Claude Fable 5 [max]	70%	$21.63	119k	88
4	GPT-5.6 Terra [max]	70%	$4.95	72k	76
5	Kimi K3 [max]	69%	$4.65	82k	98
6	GPT-5.6 Luna [max]	67%	$3.03	73k	102
7	GPT-5.5 [xhigh]	67%	$7.23	46k	82
8	Grok 4.6 [xhigh]	67%	$5.50	71k	87
9	Gemini 3.7 Flash [high]	65%	$2.18	107k	125
10	DeepSeek V4-Pro [max]	63%	$0.24	106k	155
11	Claude Opus 4.8 [max]	59%	$13.22	135k	120
12	Qwen3.8-Max [xhigh]	58%	$3.73	95k	111
13	Muse Spark 1.2 [xhigh]	55%	$3.70	99k	101
14	Claude Sonnet 5 [max]	54%	$26.40	214k	268
15	Grok 4.5 [high]	54%	$2.42	36k	61
16	DeepSeek V4-Flash [max]	53%	$0.10	108k	153
17	Muse Spark 1.1 [xhigh]	53%	$2.36	74k	96
18	GPT-5.4 [xhigh]	52%	$5.65	71k	70
19	Gemini 3.6 Flash [high]	47%	$4.42	96k	117
20	GLM 5.2 [max]	44%	$3.92	78k	129
21	Gemini 3.5 Flash [high]	36%	$3.45	76k	105
22	Kimi K2.7 Code	31%	$2.82	59k	149
23	Claude Sonnet 4.6 [high]	30%	$5.52	76k	134
24	Gemini 3.1 Pro [high]	12%	$2.14	28k	76

DeepSWE “Best” picks the highest published reasoning effort per model (not the highest pass rate). Small gaps may not be statistically meaningful — confirm on deepswe.datacurve.ai.

Head to head

Direct comparisons

Pairs a buyer would realistically weigh against each other: same tier, or a step up and down within one provider.

GPT-5.6 Sol vs GPT-5.6 Terra

GPT-5.6 Luna vs GPT-5.6 Sol

Claude Opus 5 vs GPT-5.6 Sol

Claude Fable 5 vs GPT-5.6 Sol

Gemini 3.1 Pro vs GPT-5.6 Sol

GPT-5.6 Sol vs Kimi K3

GLM 5.2 vs GPT-5.6 Sol

GLM 5.3 vs GPT-5.6 Sol

GPT-5.6 Sol vs Grok 4.6

GPT-5.6 Sol vs Qwen3.8-Max

GPT-5.6 Sol vs Qwen3.8-2.4T-A95B

GPT-5.6 Sol vs Muse Spark 1.2

GPT-5.6 Luna vs GPT-5.6 Terra

Claude Sonnet 5 vs GPT-5.6 Terra

GPT-5.6 Terra vs Grok 4.5

DeepSeek V4-Pro vs GPT-5.6 Terra

GPT-5.6 Terra vs Qwen3.8-27B

GPT-5.6 Terra vs Muse Glimmer 30B

Gemini 3.6 Flash vs GPT-5.6 Luna

Gemini 3.7 Flash vs GPT-5.6 Luna

Gemini 3.5 Flash-Lite vs GPT-5.6 Luna

GPT-5.6 Luna vs Kimi K2.6

Claude Haiku 4.5 vs GPT-5.6 Luna

DeepSeek V4-Flash vs GPT-5.6 Luna

GPT-5.6 Luna vs Ling 3.0 Flash

Claude Fable 5 vs Claude Opus 5

Claude Opus 5 vs Claude Sonnet 5

Claude Opus 5 vs Gemini 3.1 Pro

Claude Opus 5 vs Kimi K3

Claude Opus 5 vs GLM 5.2

Claude Opus 5 vs GLM 5.3

Claude Opus 5 vs Grok 4.6

Claude Haiku 4.5 vs Claude Opus 5

Claude Opus 5 vs Qwen3.8-Max

Claude Opus 5 vs Qwen3.8-2.4T-A95B

Claude Opus 5 vs Muse Spark 1.2

Claude Fable 5 vs Claude Sonnet 5

Claude Fable 5 vs Gemini 3.1 Pro

Claude Fable 5 vs Kimi K3

Claude Fable 5 vs GLM 5.2

Claude Fable 5 vs GLM 5.3

Claude Fable 5 vs Grok 4.6

Claude Fable 5 vs Claude Haiku 4.5

Claude Fable 5 vs Qwen3.8-Max

Claude Fable 5 vs Qwen3.8-2.4T-A95B

Claude Fable 5 vs Muse Spark 1.2

Claude Sonnet 5 vs Grok 4.5

Claude Haiku 4.5 vs Claude Sonnet 5

Claude Sonnet 5 vs DeepSeek V4-Pro

Claude Sonnet 5 vs Qwen3.8-27B

Claude Sonnet 5 vs Muse Glimmer 30B

Gemini 3.1 Pro vs Gemini 3.6 Flash

Gemini 3.1 Pro vs Gemini 3.7 Flash

Gemini 3.1 Pro vs Gemini 3.5 Flash-Lite

Gemini 3.1 Pro vs Kimi K3

Gemini 3.1 Pro vs GLM 5.2

Gemini 3.1 Pro vs GLM 5.3

Gemini 3.1 Pro vs Grok 4.6

Gemini 3.1 Pro vs Qwen3.8-Max

Gemini 3.1 Pro vs Qwen3.8-2.4T-A95B

Gemini 3.1 Pro vs Muse Spark 1.2

Gemini 3.6 Flash vs Gemini 3.7 Flash

Gemini 3.5 Flash-Lite vs Gemini 3.6 Flash

Gemini 3.6 Flash vs Kimi K2.6

Claude Haiku 4.5 vs Gemini 3.6 Flash

DeepSeek V4-Flash vs Gemini 3.6 Flash

Gemini 3.6 Flash vs Ling 3.0 Flash

Gemini 3.5 Flash-Lite vs Gemini 3.7 Flash

Gemini 3.7 Flash vs Kimi K2.6

Claude Haiku 4.5 vs Gemini 3.7 Flash

DeepSeek V4-Flash vs Gemini 3.7 Flash

Gemini 3.7 Flash vs Ling 3.0 Flash

Gemini 3.5 Flash-Lite vs Kimi K2.6

Claude Haiku 4.5 vs Gemini 3.5 Flash-Lite

DeepSeek V4-Flash vs Gemini 3.5 Flash-Lite

Gemini 3.5 Flash-Lite vs Ling 3.0 Flash

Kimi K2.6 vs Kimi K3

GLM 5.2 vs Kimi K3

GLM 5.3 vs Kimi K3

Grok 4.6 vs Kimi K3

Kimi K3 vs Qwen3.8-Max

Kimi K3 vs Qwen3.8-2.4T-A95B

Kimi K3 vs Muse Spark 1.2

Claude Haiku 4.5 vs Kimi K2.6

DeepSeek V4-Flash vs Kimi K2.6

Kimi K2.6 vs Ling 3.0 Flash

GLM 5.2 vs GLM 5.3

GLM 5.2 vs Grok 4.6

GLM 5.2 vs Qwen3.8-Max

GLM 5.2 vs Qwen3.8-2.4T-A95B

GLM 5.2 vs Muse Spark 1.2

GLM 5.3 vs Grok 4.6

GLM 5.3 vs Qwen3.8-Max

GLM 5.3 vs Qwen3.8-2.4T-A95B

GLM 5.3 vs Muse Spark 1.2

Grok 4.5 vs Grok 4.6

DeepSeek V4-Pro vs Grok 4.5

Grok 4.5 vs Qwen3.8-27B

Grok 4.5 vs Muse Glimmer 30B

Grok 4.6 vs Qwen3.8-Max

Grok 4.6 vs Qwen3.8-2.4T-A95B

Grok 4.6 vs Muse Spark 1.2

Claude Haiku 4.5 vs DeepSeek V4-Flash

Claude Haiku 4.5 vs Ling 3.0 Flash

DeepSeek V4-Flash vs DeepSeek V4-Pro

DeepSeek V4-Flash vs Ling 3.0 Flash

DeepSeek V4-Pro vs Qwen3.8-27B

DeepSeek V4-Pro vs Muse Glimmer 30B

Qwen3.8-2.4T-A95B vs Qwen3.8-Max

Qwen3.8-27B vs Qwen3.8-Max

Muse Spark 1.2 vs Qwen3.8-Max

Qwen3.8-2.4T-A95B vs Qwen3.8-27B

Muse Spark 1.2 vs Qwen3.8-2.4T-A95B

Muse Glimmer 30B vs Qwen3.8-27B

Muse Glimmer 30B vs Muse Spark 1.2

Claude Opus 5 vs Gemini 3.6 Flash

Claude Opus 5 vs Gemini 3.7 Flash

Claude Opus 5 vs DeepSeek V4-Flash

Claude Opus 5 vs GPT-5.6 Luna

DeepSeek V4-Flash vs GPT-5.6 Sol

Gemini 3.7 Flash vs GPT-5.6 Sol

GLM 5.2 vs Grok 4.5

GLM 5.3 vs Grok 4.5

How these comparisons are built

Every page renders from the same checked catalog dataset rather than a written template.

A comparison only exists where the underlying numbers differ. Figures we could not confirm from the provider are shown as Not verified rather than estimated, because a wrong price is more damaging than a missing one.

Benchmark results carry the source and the month they were measured. Scores in this category move with every release, and a number without a date is not much use.

Picking a tier

The tier usually matters more than the brand.

FrontierBalancedBudget

Frontier tiers earn their price on hard reasoning and long agentic tasks. For classification, extraction and summarization — most production volume — a budget tier is usually indistinguishable in output and several times cheaper.

Output tokens cost three to five times input across almost every provider here, so generation length drives the bill far more than prompt size.