Model comparisons
Head-to-head questions between two models in the catalogue, answered on price and specification rather than on quality claims we have not measured.
113 questions
Kimi K2.6 vs Kimi K3
In context →- Which is cheaper, Kimi K2.6 or Kimi K3?
- Kimi K2.6, at 100M input and 20M output tokens a month: $101.00 against $365.00, a difference of $264.00 — 72% less. That holds at any ratio of input to output on this pairing, because Kimi K2.6 has the lower rate on every column, not just the one the headline quotes. What the arithmetic does not tell you is whether both models do your job to the same standard; that is a measurement on your own prompts, not a number we can publish.
- Can I call both Kimi K2.6 and Kimi K3 today?
- Yes — both. Requests naming `moonshotai/kimi-k2.6` or `moonshotai/kimi-k3` reach a live upstream today and are billed at the rates above. 7 of the 24 models in this catalogue are callable right now; the rest are listed at their intended price and marked, so you find out here rather than from an error at call time.
- How much context does Kimi K2.6 hold compared with Kimi K3?
- Kimi K3 takes 524,288 tokens (512K) in a single request against Kimi K2.6's 262,144 (256K) — 2.0× wider. The window is a cost decision before it is a capability one: an input that does not fit has to be split, summarised in pieces and summarised again, which bills the source text more than once. Below that line the wider window is capacity you are not using.
- What licences are Kimi K2.6 and Kimi K3 under?
- Both are Modified MIT — Kimi K2.6 from Moonshot AI and Kimi K3 from Moonshot AI. Licence is therefore not a differentiator on this pairing, which is worth knowing before it becomes the question a legal review opens with. Read the text itself before shipping either; a licence name is a pointer to terms, not a summary of them.
- How do I switch between Kimi K2.6 and Kimi K3?
- Change the model id and nothing else. Both are served from `https://router.xark.io/api/v1` behind the same OpenAI-compatible contract — `moonshotai/kimi-k2.6` and `moonshotai/kimi-k3` — so the request shape, the streaming frames, the usage block and the error envelope are identical between them. Send an explicit id and no substitution happens underneath you, which is what makes running both and comparing the two on your own traffic a change of one string rather than a migration.
- How do these rates compare with the official ones?
- Both rows are published side by side rather than as a single discount claim. Kimi K2.6 is 42% below Moonshot AI's official rate and Kimi K3 is 40% below Moonshot AI's, which at 100M input and 20M output tokens a month is $101.00 against an official $175.00 for Kimi K2.6, and $365.00 against $600.00 for Kimi K3. The claim on this site is only that our rate sits below the model publisher's own — not that no one else is cheaper. Where a third-party provider undercuts us on either model, that provider's rate is printed on that model's own page.
Kimi K2.6 vs GLM-5.2
In context →- Which is cheaper, Kimi K2.6 or GLM-5.2?
- Kimi K2.6, at 100M input and 20M output tokens a month: $101.00 against $133.00, a difference of $32.00 — 24% less. That holds at any ratio of input to output on this pairing, because Kimi K2.6 has the lower rate on every column, not just the one the headline quotes. What the arithmetic does not tell you is whether both models do your job to the same standard; that is a measurement on your own prompts, not a number we can publish.
- Can I call both Kimi K2.6 and GLM-5.2 today?
- Yes — both. Requests naming `moonshotai/kimi-k2.6` or `z-ai/glm-5.2` reach a live upstream today and are billed at the rates above. 7 of the 24 models in this catalogue are callable right now; the rest are listed at their intended price and marked, so you find out here rather than from an error at call time.
- How much context does Kimi K2.6 hold compared with GLM-5.2?
- Kimi K2.6 takes 262,144 tokens (256K) in a single request against GLM-5.2's 204,800 (200K) — 1.3× wider. The window is a cost decision before it is a capability one: an input that does not fit has to be split, summarised in pieces and summarised again, which bills the source text more than once. Below that line the wider window is capacity you are not using.
- What licences are Kimi K2.6 and GLM-5.2 under?
- Kimi K2.6 is published by Moonshot AI under Modified MIT and GLM-5.2 by Z.ai under MIT. Where one of those is a plain MIT or Apache 2.0 text and the other is a publisher's own or a modified variant, the difference is real and lives in the document rather than in the name — user thresholds, field-of-use terms and attribution requirements all appear in licences of that kind. Each model's own page carries the longer answer, and the licence text carries the binding one.
- How do I switch between Kimi K2.6 and GLM-5.2?
- Change the model id and nothing else. Both are served from `https://router.xark.io/api/v1` behind the same OpenAI-compatible contract — `moonshotai/kimi-k2.6` and `z-ai/glm-5.2` — so the request shape, the streaming frames, the usage block and the error envelope are identical between them. Send an explicit id and no substitution happens underneath you, which is what makes running both and comparing the two on your own traffic a change of one string rather than a migration.
Kimi K2.6 vs DeepSeek V4 Pro
In context →- Which is cheaper, Kimi K2.6 or DeepSeek V4 Pro?
- DeepSeek V4 Pro, at 100M input and 20M output tokens a month: $39.00 against $101.00, a difference of $62.00 — 61% less. That holds at any ratio of input to output on this pairing, because DeepSeek V4 Pro has the lower rate on every column, not just the one the headline quotes. What the arithmetic does not tell you is whether both models do your job to the same standard; that is a measurement on your own prompts, not a number we can publish.
- Can I call both Kimi K2.6 and DeepSeek V4 Pro today?
- Yes — both. Requests naming `moonshotai/kimi-k2.6` or `deepseek/deepseek-v4-pro` reach a live upstream today and are billed at the rates above. 7 of the 24 models in this catalogue are callable right now; the rest are listed at their intended price and marked, so you find out here rather than from an error at call time.
- How much context does Kimi K2.6 hold compared with DeepSeek V4 Pro?
- Kimi K2.6 takes 262,144 tokens (256K) in a single request against DeepSeek V4 Pro's 163,840 (160K) — 1.6× wider. The window is a cost decision before it is a capability one: an input that does not fit has to be split, summarised in pieces and summarised again, which bills the source text more than once. Below that line the wider window is capacity you are not using.
- What licences are Kimi K2.6 and DeepSeek V4 Pro under?
- Kimi K2.6 is published by Moonshot AI under Modified MIT and DeepSeek V4 Pro by DeepSeek under MIT. Where one of those is a plain MIT or Apache 2.0 text and the other is a publisher's own or a modified variant, the difference is real and lives in the document rather than in the name — user thresholds, field-of-use terms and attribution requirements all appear in licences of that kind. Each model's own page carries the longer answer, and the licence text carries the binding one.
- How do I switch between Kimi K2.6 and DeepSeek V4 Pro?
- Change the model id and nothing else. Both are served from `https://router.xark.io/api/v1` behind the same OpenAI-compatible contract — `moonshotai/kimi-k2.6` and `deepseek/deepseek-v4-pro` — so the request shape, the streaming frames, the usage block and the error envelope are identical between them. Send an explicit id and no substitution happens underneath you, which is what makes running both and comparing the two on your own traffic a change of one string rather than a migration.
Kimi K2.6 vs DeepSeek V4 Flash
In context →- Which is cheaper, Kimi K2.6 or DeepSeek V4 Flash?
- DeepSeek V4 Flash, at 100M input and 20M output tokens a month: $12.60 against $101.00, a difference of $88.40 — 88% less. That holds at any ratio of input to output on this pairing, because DeepSeek V4 Flash has the lower rate on every column, not just the one the headline quotes. What the arithmetic does not tell you is whether both models do your job to the same standard; that is a measurement on your own prompts, not a number we can publish.
- Can I call both Kimi K2.6 and DeepSeek V4 Flash today?
- Yes — both. Requests naming `moonshotai/kimi-k2.6` or `deepseek/deepseek-v4-flash` reach a live upstream today and are billed at the rates above. 7 of the 24 models in this catalogue are callable right now; the rest are listed at their intended price and marked, so you find out here rather than from an error at call time.
- How much context does Kimi K2.6 hold compared with DeepSeek V4 Flash?
- Kimi K2.6 takes 262,144 tokens (256K) in a single request against DeepSeek V4 Flash's 163,840 (160K) — 1.6× wider. The window is a cost decision before it is a capability one: an input that does not fit has to be split, summarised in pieces and summarised again, which bills the source text more than once. Below that line the wider window is capacity you are not using.
- What licences are Kimi K2.6 and DeepSeek V4 Flash under?
- Kimi K2.6 is published by Moonshot AI under Modified MIT and DeepSeek V4 Flash by DeepSeek under MIT. Where one of those is a plain MIT or Apache 2.0 text and the other is a publisher's own or a modified variant, the difference is real and lives in the document rather than in the name — user thresholds, field-of-use terms and attribution requirements all appear in licences of that kind. Each model's own page carries the longer answer, and the licence text carries the binding one.
- How do I switch between Kimi K2.6 and DeepSeek V4 Flash?
- Change the model id and nothing else. Both are served from `https://router.xark.io/api/v1` behind the same OpenAI-compatible contract — `moonshotai/kimi-k2.6` and `deepseek/deepseek-v4-flash` — so the request shape, the streaming frames, the usage block and the error envelope are identical between them. Send an explicit id and no substitution happens underneath you, which is what makes running both and comparing the two on your own traffic a change of one string rather than a migration.
Kimi K3 vs GLM-5.2
In context →- Which is cheaper, Kimi K3 or GLM-5.2?
- GLM-5.2, at 100M input and 20M output tokens a month: $133.00 against $365.00, a difference of $232.00 — 64% less. That holds at any ratio of input to output on this pairing, because GLM-5.2 has the lower rate on every column, not just the one the headline quotes. What the arithmetic does not tell you is whether both models do your job to the same standard; that is a measurement on your own prompts, not a number we can publish.
- Can I call both Kimi K3 and GLM-5.2 today?
- Yes — both. Requests naming `moonshotai/kimi-k3` or `z-ai/glm-5.2` reach a live upstream today and are billed at the rates above. 7 of the 24 models in this catalogue are callable right now; the rest are listed at their intended price and marked, so you find out here rather than from an error at call time.
- How much context does Kimi K3 hold compared with GLM-5.2?
- Kimi K3 takes 524,288 tokens (512K) in a single request against GLM-5.2's 204,800 (200K) — 2.6× wider. The window is a cost decision before it is a capability one: an input that does not fit has to be split, summarised in pieces and summarised again, which bills the source text more than once. Below that line the wider window is capacity you are not using.
- What licences are Kimi K3 and GLM-5.2 under?
- Kimi K3 is published by Moonshot AI under Modified MIT and GLM-5.2 by Z.ai under MIT. Where one of those is a plain MIT or Apache 2.0 text and the other is a publisher's own or a modified variant, the difference is real and lives in the document rather than in the name — user thresholds, field-of-use terms and attribution requirements all appear in licences of that kind. Each model's own page carries the longer answer, and the licence text carries the binding one.
- How do I switch between Kimi K3 and GLM-5.2?
- Change the model id and nothing else. Both are served from `https://router.xark.io/api/v1` behind the same OpenAI-compatible contract — `moonshotai/kimi-k3` and `z-ai/glm-5.2` — so the request shape, the streaming frames, the usage block and the error envelope are identical between them. Send an explicit id and no substitution happens underneath you, which is what makes running both and comparing the two on your own traffic a change of one string rather than a migration.
Kimi K3 vs DeepSeek V4 Pro
In context →- Which is cheaper, Kimi K3 or DeepSeek V4 Pro?
- DeepSeek V4 Pro, at 100M input and 20M output tokens a month: $39.00 against $365.00, a difference of $326.00 — 89% less. That holds at any ratio of input to output on this pairing, because DeepSeek V4 Pro has the lower rate on every column, not just the one the headline quotes. What the arithmetic does not tell you is whether both models do your job to the same standard; that is a measurement on your own prompts, not a number we can publish.
- Can I call both Kimi K3 and DeepSeek V4 Pro today?
- Yes — both. Requests naming `moonshotai/kimi-k3` or `deepseek/deepseek-v4-pro` reach a live upstream today and are billed at the rates above. 7 of the 24 models in this catalogue are callable right now; the rest are listed at their intended price and marked, so you find out here rather than from an error at call time.
- How much context does Kimi K3 hold compared with DeepSeek V4 Pro?
- Kimi K3 takes 524,288 tokens (512K) in a single request against DeepSeek V4 Pro's 163,840 (160K) — 3.2× wider. The window is a cost decision before it is a capability one: an input that does not fit has to be split, summarised in pieces and summarised again, which bills the source text more than once. Below that line the wider window is capacity you are not using.
- What licences are Kimi K3 and DeepSeek V4 Pro under?
- Kimi K3 is published by Moonshot AI under Modified MIT and DeepSeek V4 Pro by DeepSeek under MIT. Where one of those is a plain MIT or Apache 2.0 text and the other is a publisher's own or a modified variant, the difference is real and lives in the document rather than in the name — user thresholds, field-of-use terms and attribution requirements all appear in licences of that kind. Each model's own page carries the longer answer, and the licence text carries the binding one.
- How do I switch between Kimi K3 and DeepSeek V4 Pro?
- Change the model id and nothing else. Both are served from `https://router.xark.io/api/v1` behind the same OpenAI-compatible contract — `moonshotai/kimi-k3` and `deepseek/deepseek-v4-pro` — so the request shape, the streaming frames, the usage block and the error envelope are identical between them. Send an explicit id and no substitution happens underneath you, which is what makes running both and comparing the two on your own traffic a change of one string rather than a migration.
Kimi K3 vs DeepSeek V4 Flash
In context →- Which is cheaper, Kimi K3 or DeepSeek V4 Flash?
- DeepSeek V4 Flash, at 100M input and 20M output tokens a month: $12.60 against $365.00, a difference of $352.40 — 97% less. That holds at any ratio of input to output on this pairing, because DeepSeek V4 Flash has the lower rate on every column, not just the one the headline quotes. What the arithmetic does not tell you is whether both models do your job to the same standard; that is a measurement on your own prompts, not a number we can publish.
- Can I call both Kimi K3 and DeepSeek V4 Flash today?
- Yes — both. Requests naming `moonshotai/kimi-k3` or `deepseek/deepseek-v4-flash` reach a live upstream today and are billed at the rates above. 7 of the 24 models in this catalogue are callable right now; the rest are listed at their intended price and marked, so you find out here rather than from an error at call time.
- How much context does Kimi K3 hold compared with DeepSeek V4 Flash?
- Kimi K3 takes 524,288 tokens (512K) in a single request against DeepSeek V4 Flash's 163,840 (160K) — 3.2× wider. The window is a cost decision before it is a capability one: an input that does not fit has to be split, summarised in pieces and summarised again, which bills the source text more than once. Below that line the wider window is capacity you are not using.
- What licences are Kimi K3 and DeepSeek V4 Flash under?
- Kimi K3 is published by Moonshot AI under Modified MIT and DeepSeek V4 Flash by DeepSeek under MIT. Where one of those is a plain MIT or Apache 2.0 text and the other is a publisher's own or a modified variant, the difference is real and lives in the document rather than in the name — user thresholds, field-of-use terms and attribution requirements all appear in licences of that kind. Each model's own page carries the longer answer, and the licence text carries the binding one.
- How do I switch between Kimi K3 and DeepSeek V4 Flash?
- Change the model id and nothing else. Both are served from `https://router.xark.io/api/v1` behind the same OpenAI-compatible contract — `moonshotai/kimi-k3` and `deepseek/deepseek-v4-flash` — so the request shape, the streaming frames, the usage block and the error envelope are identical between them. Send an explicit id and no substitution happens underneath you, which is what makes running both and comparing the two on your own traffic a change of one string rather than a migration.
GLM-5.2 vs DeepSeek V4 Pro
In context →- Which is cheaper, GLM-5.2 or DeepSeek V4 Pro?
- DeepSeek V4 Pro, at 100M input and 20M output tokens a month: $39.00 against $133.00, a difference of $94.00 — 71% less. That holds at any ratio of input to output on this pairing, because DeepSeek V4 Pro has the lower rate on every column, not just the one the headline quotes. What the arithmetic does not tell you is whether both models do your job to the same standard; that is a measurement on your own prompts, not a number we can publish.
- Can I call both GLM-5.2 and DeepSeek V4 Pro today?
- Yes — both. Requests naming `z-ai/glm-5.2` or `deepseek/deepseek-v4-pro` reach a live upstream today and are billed at the rates above. 7 of the 24 models in this catalogue are callable right now; the rest are listed at their intended price and marked, so you find out here rather than from an error at call time.
- How much context does GLM-5.2 hold compared with DeepSeek V4 Pro?
- GLM-5.2 takes 204,800 tokens (200K) in a single request against DeepSeek V4 Pro's 163,840 (160K) — 1.3× wider. The window is a cost decision before it is a capability one: an input that does not fit has to be split, summarised in pieces and summarised again, which bills the source text more than once. Below that line the wider window is capacity you are not using.
- What licences are GLM-5.2 and DeepSeek V4 Pro under?
- Both are MIT — GLM-5.2 from Z.ai and DeepSeek V4 Pro from DeepSeek. Licence is therefore not a differentiator on this pairing, which is worth knowing before it becomes the question a legal review opens with. Read the text itself before shipping either; a licence name is a pointer to terms, not a summary of them.
- How do I switch between GLM-5.2 and DeepSeek V4 Pro?
- Change the model id and nothing else. Both are served from `https://router.xark.io/api/v1` behind the same OpenAI-compatible contract — `z-ai/glm-5.2` and `deepseek/deepseek-v4-pro` — so the request shape, the streaming frames, the usage block and the error envelope are identical between them. Send an explicit id and no substitution happens underneath you, which is what makes running both and comparing the two on your own traffic a change of one string rather than a migration.
The rest, answered on their own pages
Every question below is answered in full where it belongs, beside the rate table and the specification it refers to.
GLM-5.2 Air vs DeepSeek V4 Pro
- Which is cheaper, GLM-5.2 Air or DeepSeek V4 Pro?
- Can I call both GLM-5.2 Air and DeepSeek V4 Pro today?
- How much context does GLM-5.2 Air hold compared with DeepSeek V4 Pro?
- What licences are GLM-5.2 Air and DeepSeek V4 Pro under?
- How do I switch between GLM-5.2 Air and DeepSeek V4 Pro?
- At what workload does DeepSeek V4 Pro become the cheaper of the two?
DeepSeek V4 Pro vs Qwen3 Max Instruct
- Which is cheaper, DeepSeek V4 Pro or Qwen3 Max Instruct?
- Can I call both DeepSeek V4 Pro and Qwen3 Max Instruct today?
- How much context does DeepSeek V4 Pro hold compared with Qwen3 Max Instruct?
- What licences are DeepSeek V4 Pro and Qwen3 Max Instruct under?
- How do I switch between DeepSeek V4 Pro and Qwen3 Max Instruct?
DeepSeek V4 Flash vs Qwen3 Max Instruct
- Which is cheaper, DeepSeek V4 Flash or Qwen3 Max Instruct?
- Can I call both DeepSeek V4 Flash and Qwen3 Max Instruct today?
- How much context does DeepSeek V4 Flash hold compared with Qwen3 Max Instruct?
- What licences are DeepSeek V4 Flash and Qwen3 Max Instruct under?
- How do I switch between DeepSeek V4 Flash and Qwen3 Max Instruct?
Check it against your own numbers
$5 in free credits, no card required.
Every rate quoted above is the rate the API bills.