mirror of
https://github.com/Heretek-AI/Token-Max.git
synced 2026-10-12 02:03:18 +00:00
feat(plans): recalibrate OpenCode Go with 28-model usage limits, allowances, and 5h/7d caps
This commit is contained in:
John Smith
@@ -460,9 +460,9 @@ Every tier in Token-Max is classified with explicit provenance in `estimateMeta`
|
||||
|
||||
### OpenCode (`opencode`)
|
||||
|
||||
- **Primary Pricing & Docs:** [https://opencode.ai/zen](https://opencode.ai/zen)
|
||||
- **Primary Pricing & Docs:** [https://opencode.ai/go](https://opencode.ai/go) · Usage Limits: [https://opencode.ai/docs/go/#usage-limits](https://opencode.ai/docs/go/#usage-limits) · Zen PAYG: [https://opencode.ai/zen](https://opencode.ai/zen)
|
||||
- **Category:** `coding-router`
|
||||
- **Data Privacy & Training:** Zero-retention; no training on Zen
|
||||
- **Data Privacy & Training:** Zero data retention (0 days) on most Go models; 30 days on Grok 4.6 & GPT-5.6 Luna; Muse Spark Contributor trains on code. Zen models US-hosted with zero retention
|
||||
- **IP Indemnity:** False
|
||||
- **Stacking Policy:** `prohibited` — *"Terms of Service: prohibits users who "create, maintain, or use multiple accounts to circumvent usage limits, access restrictions, billing obligations... or any other restriction or policy.""*
|
||||
|
||||
@@ -472,13 +472,15 @@ Every tier in Token-Max is classified with explicit provenance in `estimateMeta`
|
||||
| :--- | :--- | :--- | :--- | :--- | :--- |
|
||||
| **OpenSource CLI** | Free ($0) | **usage**: Free open-source agent; BYOK to any provider | BYOK any provider | **Token cost = your provider's rates**<br/>Floor: `0M` | [RESEARCH (low)](https://opencode.ai/zen) |
|
||||
| **Zen (PAYG)** | $20/mo | **billing**: Pay-as-you-go dollar balance, min $20 top-up (+$1.23 card fee)<br/>**markup**: Zero markup per request<br/>**autoTopUp**: $20 when balance < $5 | GPT-5.6 Luna, GPT-5.6 Sol, Claude Fable 5.1<br/>*(+5 more)* | **Prepaid balance (~10M fee-equivalent tokens)**<br/>Floor: `10M` | Mid: `20M` | Opt: `40M` | [RESEARCH (low)](https://opencode.ai/zen) |
|
||||
| **Go** | $10/mo | **usage**: Open-model subscription ($10/mo) for stable global access | Qwen3.7 Plus, Kimi K3, Kimi K2.7<br/>*(+2 more)* | **Open-model subscription (~30M tokens/mo est.)**<br/>Floor: `30M` | [RESEARCH (low)](https://opencode.ai/zen) |
|
||||
| **Go** | $10/mo | **monthlyAllowance**: Per-model monthly usage: $60 (tier 1 models), $30 (tier 2), $15 (tier 3)<br/>**fiveHourCap**: 20% of monthly allowance in any 5h ($12 / $6 / $3)<br/>**weeklyCap**: 50% of monthly allowance in any 7d ($30 / $15 / $7.50)<br/>**monthlyCap**: 100% of monthly allowance ($60 / $30 / $15)<br/>**requests**: Up to 31,580/mo on GLM-5.3-Flash, 65,000/mo on DeepSeek Flash, 150,400/mo on MiMo-V2.5, 226,600/mo on Muse Spark<br/>**topUp**: Optional 'Use balance' falls back to Zen PAYG balance when limits reached | Kimi K3, Qwen 3.8 Max, Grok 4.6, GLM-5.3, GLM-5.2, DeepSeek V4.1 Flash, MiniMax M3, MiMo-V2.5, Muse Spark<br/>*(28 models)* | **$10/mo open-model subscription ($15–$60 per-model allowances; ~30M–752M tokens)**<br/>Floor: `30M` | Mid: `95M` | Opt: `752M` | [OFFICIAL (high)](https://opencode.ai/docs/go/#usage-limits)<br/>*"Usage limits are defined as monthly dollar amounts... Each model has the following usage limits: 5-hour — 20% of the monthly limit; weekly — 50%; and monthly — 100%. For example, if a model has a $60 monthly limit, you can spend up to: 5-hour limit — $12, weekly limit — $30, monthly limit — $60."* |
|
||||
|
||||
**Key Gotchas & Constraints:**
|
||||
- The CLI itself is free/open-source; opencode.ai/pricing is retired (404) - offerings are Zen (PAYG) and Go ($10/mo)
|
||||
- Zen charges $1.23 card processing fee on top-ups; auto top-up at $20 below $5 balance
|
||||
- GPT-5.6 Sol carries 50% discount through Sept 18, 2026
|
||||
- Zen models hosted in US with zero-retention provider policy
|
||||
- OpenCode Go ($10/mo) provides $15, $30, or $60 monthly usage allowance per model with 20% 5h and 50% weekly caps
|
||||
- Optional 'Use balance' falls back to Zen PAYG balance when Go model limits are reached instead of blocking requests
|
||||
- DeepSeek V4.1 Flash has 4× promotional limits ($60/mo limit; 26K req / 5h) through Sep 20, 2026 (normally $15)
|
||||
- Muse Spark Contributor models train on prompt/output data and are not zero data retention (ZDR); all other Go models provide zero training (0 days retention, except Grok 4.6 and GPT-5.6 Luna with 30 days retention)
|
||||
- Zen PAYG charges $1.23 card processing fee on top-ups; auto top-up at $20 below $5 balance
|
||||
- The CLI itself is free/open-source; opencode.ai/pricing is retired (404) - offerings are Go ($10/mo) and Zen (PAYG)
|
||||
|
||||
---
|
||||
|
||||
|
||||
+204
-28
@@ -2,7 +2,7 @@
|
||||
"id": "opencode",
|
||||
"name": "OpenCode",
|
||||
"category": "coding-router",
|
||||
"url": "https://opencode.ai/zen",
|
||||
"url": "https://opencode.ai/go",
|
||||
"lastVerified": "2026-09-19",
|
||||
"tiers": [
|
||||
{
|
||||
@@ -19,7 +19,7 @@
|
||||
"estimatedMillionTokens": 0,
|
||||
"assumptions": "No fixed bundle; you pay your API provider directly",
|
||||
"estimateMeta": {
|
||||
"sourceUrl": "https://opencode.ai/zen",
|
||||
"sourceUrl": "https://opencode.ai/go",
|
||||
"sourceType": "research",
|
||||
"confidence": "low",
|
||||
"verifiedAt": "2026-09-19"
|
||||
@@ -52,7 +52,7 @@
|
||||
"optimisticEstimate": 40,
|
||||
"assumptions": "$20 minimum balance at zero markup buys $20 of inference at curated gateway rates; conservative 10M at ~$2.00/M frontier blend, midpoint 20M at ~$1.00/M, optimistic 40M at ~$0.50/M open-model rates; excludes the $1.23 card fee",
|
||||
"estimateMeta": {
|
||||
"sourceUrl": "https://opencode.ai/zen",
|
||||
"sourceUrl": "https://opencode.ai/go",
|
||||
"sourceType": "research",
|
||||
"confidence": "low",
|
||||
"verifiedAt": "2026-09-19"
|
||||
@@ -64,45 +64,221 @@
|
||||
"name": "Go",
|
||||
"monthlyPrice": 10,
|
||||
"limits": {
|
||||
"usage": "Open-model subscription ($10/mo) for stable global access"
|
||||
"monthlyAllowance": "Per-model monthly usage: $60 (tier 1 models), $30 (tier 2), $15 (tier 3)",
|
||||
"fiveHourCap": "20% of monthly allowance in any 5h ($12 / $6 / $3)",
|
||||
"weeklyCap": "50% of monthly allowance in any 7d ($30 / $15 / $7.50)",
|
||||
"monthlyCap": "100% of monthly allowance ($60 / $30 / $15)",
|
||||
"requests": "Up to 31,580/mo on GLM-5.3-Flash, 65,000/mo on DeepSeek Flash, 150,400/mo on MiMo-V2.5",
|
||||
"topUp": "Optional 'Use balance' falls back to Zen PAYG balance when limits reached"
|
||||
},
|
||||
"models": [
|
||||
"Qwen3.7 Plus",
|
||||
"Kimi K3",
|
||||
"Kimi K2.7",
|
||||
"Qwen 3.8 Max",
|
||||
"Grok 4.6",
|
||||
"Qwen 3.7 Max",
|
||||
"GLM-5.3",
|
||||
"GLM-5.2",
|
||||
"GLM-5.1",
|
||||
"DeepSeek V4 Pro",
|
||||
"Kimi K2.6",
|
||||
"Kimi K2.7 Code",
|
||||
"Hy4 preview",
|
||||
"GPT-5.6 Luna",
|
||||
"MiMo-V2.5"
|
||||
"MiniMax M3",
|
||||
"MiMo-V2.5-Pro",
|
||||
"Qwen 3.6 Plus",
|
||||
"MiniMax M2.7",
|
||||
"MiniMax M2.5",
|
||||
"Qwen 3.7 Plus",
|
||||
"Hy3",
|
||||
"Qwen 3.8 Flash",
|
||||
"GLM-5.3 Flash",
|
||||
"DeepSeek V4 Flash Vision Exp",
|
||||
"LongCat-2.0",
|
||||
"DeepSeek V4 Flash",
|
||||
"DeepSeek V4.1 Flash",
|
||||
"MiMo-V2.5",
|
||||
"Muse Spark 1.3 Contributor",
|
||||
"Muse Spark 1.2 Contributor"
|
||||
],
|
||||
"modelAllowances": {
|
||||
"deepseek": 60.0,
|
||||
"glm": 60.0,
|
||||
"economy": 60.0,
|
||||
"standard": 30.0,
|
||||
"frontier": 15.0,
|
||||
"tier-60": 60.0,
|
||||
"tier-30": 30.0,
|
||||
"tier-15": 15.0,
|
||||
"default": 30.0
|
||||
},
|
||||
"estimatedTokenBudget": {
|
||||
"description": "Open-model subscription (~30M tokens/mo est.)",
|
||||
"estimatedMillionTokens": 30,
|
||||
"assumptions": "$10 / ~$0.3/M blended open-model rate ~= 30M tokens; exact cap not published",
|
||||
"estimateMeta": {
|
||||
"sourceUrl": "https://opencode.ai/zen",
|
||||
"sourceType": "research",
|
||||
"confidence": "low",
|
||||
"verifiedAt": "2026-09-19"
|
||||
"perModelTokenBudgets": {
|
||||
"kimi k3": {
|
||||
"estimatedMillionTokens": 7.85,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"qwen 3.8 max": {
|
||||
"estimatedMillionTokens": 13.4,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"grok 4.6": {
|
||||
"estimatedMillionTokens": 11.56,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"qwen 3.7 max": {
|
||||
"estimatedMillionTokens": 19.57,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"glm-5.3": {
|
||||
"estimatedMillionTokens": 17.57,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"glm-5.2": {
|
||||
"estimatedMillionTokens": 70.29,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"glm-5.1": {
|
||||
"estimatedMillionTokens": 70.29,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"deepseek v4 pro": {
|
||||
"estimatedMillionTokens": 46.0,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"kimi k2.6": {
|
||||
"estimatedMillionTokens": 97.44,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"kimi k2.7 code": {
|
||||
"estimatedMillionTokens": 94.16,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"hy4 preview": {
|
||||
"estimatedMillionTokens": 71.07,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"gpt-5.6 luna": {
|
||||
"estimatedMillionTokens": 109.57,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"minimax m3": {
|
||||
"estimatedMillionTokens": 302.7,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"mimo-v2.5-pro": {
|
||||
"estimatedMillionTokens": 80.46,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"qwen 3.6 plus": {
|
||||
"estimatedMillionTokens": 175.3,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"minimax m2.7": {
|
||||
"estimatedMillionTokens": 302.7,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"minimax m2.5": {
|
||||
"estimatedMillionTokens": 302.7,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"qwen 3.7 plus": {
|
||||
"estimatedMillionTokens": 254.55,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"hy3": {
|
||||
"estimatedMillionTokens": 609.43,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"qwen 3.8 flash": {
|
||||
"estimatedMillionTokens": 357.95,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"glm-5.3 flash": {
|
||||
"estimatedMillionTokens": 635.96,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"deepseek v4 flash vision exp": {
|
||||
"estimatedMillionTokens": 187.92,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"longcat-2.0": {
|
||||
"estimatedMillionTokens": 375.84,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"deepseek v4 flash": {
|
||||
"estimatedMillionTokens": 375.84,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"deepseek v4.1 flash": {
|
||||
"estimatedMillionTokens": 751.68,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"mimo-v2.5": {
|
||||
"estimatedMillionTokens": 980.93,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"muse spark 1.3 contributor": {
|
||||
"estimatedMillionTokens": 1373.3,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"muse spark 1.2 contributor": {
|
||||
"estimatedMillionTokens": 1373.3,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
}
|
||||
},
|
||||
"notes": "New Sept 2026 low-cost subscription tier"
|
||||
"estimatedTokenBudget": {
|
||||
"description": "$10/mo open-model subscription ($15–$60 per-model allowances; ~30M–752M tokens)",
|
||||
"estimatedMillionTokens": 30,
|
||||
"midpointEstimate": 95,
|
||||
"optimisticEstimate": 752,
|
||||
"assumptions": "$10/mo plan fee provides independent monthly usage allowances per model ($15, $30, or $60) with 20% 5h and 50% weekly caps; tokens range from 7.85M (Kimi K3 at $15 allowance) to 94M (Kimi K2.7 Code at $60) up to 752M (DeepSeek V4.1 Flash at $60 promo) and 981M (MiMo-V2.5)",
|
||||
"estimateMeta": {
|
||||
"sourceUrl": "https://opencode.ai/docs/go/#usage-limits",
|
||||
"sourceQuote": "Usage limits are defined as monthly dollar amounts... Each model has the following usage limits: 5-hour — 20% of the monthly limit; weekly — 50%; and monthly — 100%. For example, if a model has a $60 monthly limit, you can spend up to: 5-hour limit — $12, weekly limit — $30, monthly limit — $60.",
|
||||
"sourceType": "official",
|
||||
"confidence": "high",
|
||||
"verifiedAt": "2026-09-19",
|
||||
"basisModel": "open-model-allowances"
|
||||
}
|
||||
},
|
||||
"notes": "New low-cost open-model subscription with independent allowances per model"
|
||||
}
|
||||
],
|
||||
"gotchas": [
|
||||
"The CLI itself is free/open-source; opencode.ai/pricing is retired (404) - offerings are Zen (PAYG) and Go ($10/mo)",
|
||||
"Zen charges $1.23 card processing fee on top-ups; auto top-up at $20 below $5 balance",
|
||||
"GPT-5.6 Sol carries 50% discount through Sept 18, 2026",
|
||||
"Zen models hosted in US with zero-retention provider policy"
|
||||
"OpenCode Go ($10/mo) provides $15, $30, or $60 monthly usage allowance per model with 20% 5h and 50% weekly caps",
|
||||
"Optional 'Use balance' falls back to Zen PAYG balance when Go model limits are reached",
|
||||
"DeepSeek V4.1 Flash has 4× promotional limits ($60/mo limit) through Sep 20, 2026 (normally $15)",
|
||||
"Muse Spark Contributor models train on prompt/output data and are not zero data retention (ZDR); all other Go models provide zero training (0 days retention, except Grok 4.6 and GPT-5.6 Luna with 30 days retention)",
|
||||
"Zen PAYG charges $1.23 card processing fee on top-ups; auto top-up at $20 below $5 balance"
|
||||
],
|
||||
"tosHighlights": [],
|
||||
"dataTraining": "Zero-retention; no training on Zen",
|
||||
"tosHighlights": [
|
||||
"Zero data retention on most Go models (0 days); 30 days on Grok 4.6 and GPT-5.6 Luna; Muse Spark Contributor trains on code",
|
||||
"Zen models US-hosted with zero retention"
|
||||
],
|
||||
"dataTraining": "Zero data retention (0 days) on most models; 30 days on Grok 4.6 & GPT-5.6 Luna; Muse Spark Contributor trains on code",
|
||||
"ipIndemnity": false,
|
||||
"stackingPolicy": "prohibited",
|
||||
"stackingPolicyNote": "Terms of Service: prohibits users who \"create, maintain, or use multiple accounts to circumvent usage limits, access restrictions, billing obligations... or any other restriction or policy.\""
|
||||
|
||||
+91
-18
@@ -2677,6 +2677,38 @@ write_json(
|
||||
},
|
||||
)
|
||||
|
||||
_OPENCODE_GO_PRICES = {
|
||||
"kimi k3": (15.0, {"input": 3.00, "output": 15.00, "cacheRead": 0.30, "cacheWrite": None}),
|
||||
"qwen 3.8 max": (15.0, {"input": 2.00, "output": 6.00, "cacheRead": 0.25, "cacheWrite": 2.50}),
|
||||
"grok 4.6": (15.0, {"input": 2.00, "output": 6.00, "cacheRead": 0.50, "cacheWrite": None}),
|
||||
"qwen 3.7 max": (30.0, {"input": 2.50, "output": 7.50, "cacheRead": 0.50, "cacheWrite": 3.125}),
|
||||
"glm-5.3": (15.0, {"input": 1.40, "output": 4.40, "cacheRead": 0.26, "cacheWrite": None}),
|
||||
"glm-5.2": (60.0, {"input": 1.40, "output": 4.40, "cacheRead": 0.26, "cacheWrite": None}),
|
||||
"glm-5.1": (60.0, {"input": 1.40, "output": 4.40, "cacheRead": 0.26, "cacheWrite": None}),
|
||||
"deepseek v4 pro": (15.0, {"input": 0.66, "output": 1.98, "cacheRead": 0.022, "cacheWrite": None}),
|
||||
"kimi k2.6": (60.0, {"input": 0.95, "output": 4.00, "cacheRead": 0.16, "cacheWrite": None}),
|
||||
"kimi k2.7 code": (60.0, {"input": 0.95, "output": 4.00, "cacheRead": 0.19, "cacheWrite": None}),
|
||||
"hy4 preview": (30.0, {"input": 0.834, "output": 2.501, "cacheRead": 0.042, "cacheWrite": None}),
|
||||
"gpt-5.6 luna": (15.0, {"input": 0.20, "output": 1.20, "cacheRead": 0.02, "cacheWrite": 0.25}),
|
||||
"minimax m3": (60.0, {"input": 0.30, "output": 1.20, "cacheRead": 0.06, "cacheWrite": None}),
|
||||
"mimo-v2.5-pro": (15.0, {"input": 0.435, "output": 0.87, "cacheRead": 0.003625, "cacheWrite": None}),
|
||||
"qwen 3.6 plus": (60.0, {"input": 0.50, "output": 3.00, "cacheRead": 0.05, "cacheWrite": 0.625}),
|
||||
"minimax m2.7": (60.0, {"input": 0.30, "output": 1.20, "cacheRead": 0.06, "cacheWrite": 0.375}),
|
||||
"minimax m2.5": (60.0, {"input": 0.30, "output": 1.20, "cacheRead": 0.06, "cacheWrite": 0.375}),
|
||||
"qwen 3.7 plus": (60.0, {"input": 0.40, "output": 1.60, "cacheRead": 0.04, "cacheWrite": 0.50}),
|
||||
"hy3": (60.0, {"input": 0.14, "output": 0.58, "cacheRead": 0.035, "cacheWrite": None}),
|
||||
"qwen 3.8 flash": (30.0, {"input": 0.15, "output": 0.47, "cacheRead": 0.016, "cacheWrite": 0.20}),
|
||||
"glm-5.3 flash": (60.0, {"input": 0.15, "output": 0.50, "cacheRead": 0.03, "cacheWrite": None}),
|
||||
"deepseek v4 flash vision exp": (15.0, {"input": 0.15, "output": 0.60, "cacheRead": 0.003, "cacheWrite": None}),
|
||||
"longcat-2.0": (60.0, {"input": 0.30, "output": 1.20, "cacheRead": 0.006, "cacheWrite": None}),
|
||||
"deepseek v4 flash": (30.0, {"input": 0.15, "output": 0.60, "cacheRead": 0.003, "cacheWrite": None}),
|
||||
"deepseek v4.1 flash": (60.0, {"input": 0.15, "output": 0.60, "cacheRead": 0.003, "cacheWrite": None}),
|
||||
"mimo-v2.5": (60.0, {"input": 0.14, "output": 0.28, "cacheRead": 0.0028, "cacheWrite": None}),
|
||||
"muse spark 1.3 contributor": (60.0, {"input": 0.10, "output": 0.20, "cacheRead": 0.002, "cacheWrite": None}),
|
||||
"muse spark 1.2 contributor": (60.0, {"input": 0.10, "output": 0.20, "cacheRead": 0.002, "cacheWrite": None}),
|
||||
}
|
||||
_OPENCODE_GO_POOL = custom_model_pool(_OPENCODE_GO_PRICES, confidence="high")
|
||||
|
||||
# 18. OpenCode
|
||||
write_json(
|
||||
"opencode.json",
|
||||
@@ -2684,7 +2716,7 @@ write_json(
|
||||
"id": "opencode",
|
||||
"name": "OpenCode",
|
||||
"category": "coding-router",
|
||||
"url": "https://opencode.ai/zen",
|
||||
"url": "https://opencode.ai/go",
|
||||
"lastVerified": LAST_VERIFIED,
|
||||
"tiers": [
|
||||
{
|
||||
@@ -2732,39 +2764,80 @@ write_json(
|
||||
"name": "Go",
|
||||
"monthlyPrice": 10,
|
||||
"limits": {
|
||||
"usage": "Open-model subscription ($10/mo) for stable global access",
|
||||
"monthlyAllowance": "Per-model monthly usage: $60 (tier 1 models), $30 (tier 2), $15 (tier 3)",
|
||||
"fiveHourCap": "20% of monthly allowance in any 5h ($12 / $6 / $3)",
|
||||
"weeklyCap": "50% of monthly allowance in any 7d ($30 / $15 / $7.50)",
|
||||
"monthlyCap": "100% of monthly allowance ($60 / $30 / $15)",
|
||||
"requests": "Up to 31,580/mo on GLM-5.3-Flash, 65,000/mo on DeepSeek Flash, 150,400/mo on MiMo-V2.5",
|
||||
"topUp": "Optional 'Use balance' falls back to Zen PAYG balance when limits reached",
|
||||
},
|
||||
"models": [
|
||||
"Qwen3.7 Plus",
|
||||
"Kimi K3",
|
||||
"Kimi K2.7",
|
||||
"Qwen 3.8 Max",
|
||||
"Grok 4.6",
|
||||
"Qwen 3.7 Max",
|
||||
"GLM-5.3",
|
||||
"GLM-5.2",
|
||||
"GLM-5.1",
|
||||
"DeepSeek V4 Pro",
|
||||
"Kimi K2.6",
|
||||
"Kimi K2.7 Code",
|
||||
"Hy4 preview",
|
||||
"GPT-5.6 Luna",
|
||||
"MiniMax M3",
|
||||
"MiMo-V2.5-Pro",
|
||||
"Qwen 3.6 Plus",
|
||||
"MiniMax M2.7",
|
||||
"MiniMax M2.5",
|
||||
"Qwen 3.7 Plus",
|
||||
"Hy3",
|
||||
"Qwen 3.8 Flash",
|
||||
"GLM-5.3 Flash",
|
||||
"DeepSeek V4 Flash Vision Exp",
|
||||
"LongCat-2.0",
|
||||
"DeepSeek V4 Flash",
|
||||
"DeepSeek V4.1 Flash",
|
||||
"MiMo-V2.5",
|
||||
"Muse Spark 1.3 Contributor",
|
||||
"Muse Spark 1.2 Contributor",
|
||||
],
|
||||
"modelAllowances": {
|
||||
"deepseek": 60.0,
|
||||
"glm": 60.0,
|
||||
"economy": 60.0,
|
||||
"standard": 30.0,
|
||||
"frontier": 15.0,
|
||||
"tier-60": 60.0,
|
||||
"tier-30": 30.0,
|
||||
"tier-15": 15.0,
|
||||
"default": 30.0,
|
||||
},
|
||||
"perModelTokenBudgets": _OPENCODE_GO_POOL,
|
||||
"estimatedTokenBudget": {
|
||||
"description": "Open-model subscription (~30M tokens/mo est.)",
|
||||
"description": "$10/mo open-model subscription ($15–$60 per-model allowances; ~30M–752M tokens)",
|
||||
"estimatedMillionTokens": 30,
|
||||
"assumptions": "$10 / ~$0.3/M blended open-model rate ~= 30M tokens; exact cap not published",
|
||||
"midpointEstimate": 95,
|
||||
"optimisticEstimate": 752,
|
||||
"assumptions": "$10/mo plan fee provides independent monthly usage allowances per model ($15, $30, or $60) with 20% 5h and 50% weekly caps; tokens range from 7.85M (Kimi K3 at $15 allowance) to 94M (Kimi K2.7 Code at $60) up to 752M (DeepSeek V4.1 Flash at $60 promo) and 981M (MiMo-V2.5)",
|
||||
"estimateMeta": {
|
||||
"sourceUrl": "https://opencode.ai/docs/go/#usage-limits",
|
||||
"sourceQuote": "Usage limits are defined as monthly dollar amounts... Each model has the following usage limits: 5-hour — 20% of the monthly limit; weekly — 50%; and monthly — 100%. For example, if a model has a $60 monthly limit, you can spend up to: 5-hour limit — $12, weekly limit — $30, monthly limit — $60.",
|
||||
"sourceType": "official",
|
||||
"confidence": "high",
|
||||
"verifiedAt": LAST_VERIFIED,
|
||||
"basisModel": "open-model-allowances",
|
||||
},
|
||||
},
|
||||
"notes": "New Sept 2026 low-cost subscription tier",
|
||||
"notes": "New low-cost open-model subscription with independent allowances per model",
|
||||
},
|
||||
],
|
||||
"gotchas": [
|
||||
"The CLI itself is free/open-source; opencode.ai/pricing is retired (404) - offerings are Zen (PAYG) and Go ($10/mo)",
|
||||
"Zen charges $1.23 card processing fee on top-ups; auto top-up at $20 below $5 balance",
|
||||
"GPT-5.6 Sol carries 50% discount through Sept 18, 2026",
|
||||
"Zen models hosted in US with zero-retention provider policy",
|
||||
"OpenCode Go ($10/mo) provides $15, $30, or $60 monthly usage allowance per model with 20% 5h and 50% weekly caps",
|
||||
"Optional 'Use balance' falls back to Zen PAYG balance when Go model limits are reached",
|
||||
"DeepSeek V4.1 Flash has 4× promotional limits ($60/mo limit) through Sep 20, 2026 (normally $15)",
|
||||
"Muse Spark Contributor models train on prompt/output data and are not zero data retention (ZDR); all other Go models provide zero training (0 days retention, except Grok 4.6 and GPT-5.6 Luna with 30 days retention)",
|
||||
"Zen PAYG charges $1.23 card processing fee on top-ups; auto top-up at $20 below $5 balance",
|
||||
],
|
||||
"tosHighlights": [],
|
||||
"dataTraining": "Zero-retention; no training on Zen",
|
||||
"tosHighlights": [
|
||||
"Zero data retention on most Go models (0 days); 30 days on Grok 4.6 and GPT-5.6 Luna; Muse Spark Contributor trains on code",
|
||||
"Zen models US-hosted with zero retention",
|
||||
],
|
||||
"dataTraining": "Zero data retention (0 days) on most models; 30 days on Grok 4.6 & GPT-5.6 Luna; Muse Spark Contributor trains on code",
|
||||
"ipIndemnity": false,
|
||||
},
|
||||
)
|
||||
|
||||
@@ -1,3 +1,3 @@
|
||||
{
|
||||
"timestamp": "2026-09-20T01:44:26.566Z"
|
||||
"timestamp": "2026-09-20T01:48:06.600Z"
|
||||
}
|
||||
+204
-28
@@ -4266,7 +4266,7 @@
|
||||
"id": "opencode",
|
||||
"name": "OpenCode",
|
||||
"category": "coding-router",
|
||||
"url": "https://opencode.ai/zen",
|
||||
"url": "https://opencode.ai/go",
|
||||
"lastVerified": "2026-09-19",
|
||||
"tiers": [
|
||||
{
|
||||
@@ -4283,7 +4283,7 @@
|
||||
"estimatedMillionTokens": 0,
|
||||
"assumptions": "No fixed bundle; you pay your API provider directly",
|
||||
"estimateMeta": {
|
||||
"sourceUrl": "https://opencode.ai/zen",
|
||||
"sourceUrl": "https://opencode.ai/go",
|
||||
"sourceType": "research",
|
||||
"confidence": "low",
|
||||
"verifiedAt": "2026-09-19"
|
||||
@@ -4316,7 +4316,7 @@
|
||||
"optimisticEstimate": 40,
|
||||
"assumptions": "$20 minimum balance at zero markup buys $20 of inference at curated gateway rates; conservative 10M at ~$2.00/M frontier blend, midpoint 20M at ~$1.00/M, optimistic 40M at ~$0.50/M open-model rates; excludes the $1.23 card fee",
|
||||
"estimateMeta": {
|
||||
"sourceUrl": "https://opencode.ai/zen",
|
||||
"sourceUrl": "https://opencode.ai/go",
|
||||
"sourceType": "research",
|
||||
"confidence": "low",
|
||||
"verifiedAt": "2026-09-19"
|
||||
@@ -4328,45 +4328,221 @@
|
||||
"name": "Go",
|
||||
"monthlyPrice": 10,
|
||||
"limits": {
|
||||
"usage": "Open-model subscription ($10/mo) for stable global access"
|
||||
"monthlyAllowance": "Per-model monthly usage: $60 (tier 1 models), $30 (tier 2), $15 (tier 3)",
|
||||
"fiveHourCap": "20% of monthly allowance in any 5h ($12 / $6 / $3)",
|
||||
"weeklyCap": "50% of monthly allowance in any 7d ($30 / $15 / $7.50)",
|
||||
"monthlyCap": "100% of monthly allowance ($60 / $30 / $15)",
|
||||
"requests": "Up to 31,580/mo on GLM-5.3-Flash, 65,000/mo on DeepSeek Flash, 150,400/mo on MiMo-V2.5",
|
||||
"topUp": "Optional 'Use balance' falls back to Zen PAYG balance when limits reached"
|
||||
},
|
||||
"models": [
|
||||
"Qwen3.7 Plus",
|
||||
"Kimi K3",
|
||||
"Kimi K2.7",
|
||||
"Qwen 3.8 Max",
|
||||
"Grok 4.6",
|
||||
"Qwen 3.7 Max",
|
||||
"GLM-5.3",
|
||||
"GLM-5.2",
|
||||
"GLM-5.1",
|
||||
"DeepSeek V4 Pro",
|
||||
"Kimi K2.6",
|
||||
"Kimi K2.7 Code",
|
||||
"Hy4 preview",
|
||||
"GPT-5.6 Luna",
|
||||
"MiMo-V2.5"
|
||||
"MiniMax M3",
|
||||
"MiMo-V2.5-Pro",
|
||||
"Qwen 3.6 Plus",
|
||||
"MiniMax M2.7",
|
||||
"MiniMax M2.5",
|
||||
"Qwen 3.7 Plus",
|
||||
"Hy3",
|
||||
"Qwen 3.8 Flash",
|
||||
"GLM-5.3 Flash",
|
||||
"DeepSeek V4 Flash Vision Exp",
|
||||
"LongCat-2.0",
|
||||
"DeepSeek V4 Flash",
|
||||
"DeepSeek V4.1 Flash",
|
||||
"MiMo-V2.5",
|
||||
"Muse Spark 1.3 Contributor",
|
||||
"Muse Spark 1.2 Contributor"
|
||||
],
|
||||
"modelAllowances": {
|
||||
"deepseek": 60,
|
||||
"glm": 60,
|
||||
"economy": 60,
|
||||
"standard": 30,
|
||||
"frontier": 15,
|
||||
"tier-60": 60,
|
||||
"tier-30": 30,
|
||||
"tier-15": 15,
|
||||
"default": 30
|
||||
},
|
||||
"estimatedTokenBudget": {
|
||||
"description": "Open-model subscription (~30M tokens/mo est.)",
|
||||
"estimatedMillionTokens": 30,
|
||||
"assumptions": "$10 / ~$0.3/M blended open-model rate ~= 30M tokens; exact cap not published",
|
||||
"estimateMeta": {
|
||||
"sourceUrl": "https://opencode.ai/zen",
|
||||
"sourceType": "research",
|
||||
"confidence": "low",
|
||||
"verifiedAt": "2026-09-19"
|
||||
"perModelTokenBudgets": {
|
||||
"kimi k3": {
|
||||
"estimatedMillionTokens": 7.85,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"qwen 3.8 max": {
|
||||
"estimatedMillionTokens": 13.4,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"grok 4.6": {
|
||||
"estimatedMillionTokens": 11.56,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"qwen 3.7 max": {
|
||||
"estimatedMillionTokens": 19.57,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"glm-5.3": {
|
||||
"estimatedMillionTokens": 17.57,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"glm-5.2": {
|
||||
"estimatedMillionTokens": 70.29,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"glm-5.1": {
|
||||
"estimatedMillionTokens": 70.29,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"deepseek v4 pro": {
|
||||
"estimatedMillionTokens": 46,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"kimi k2.6": {
|
||||
"estimatedMillionTokens": 97.44,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"kimi k2.7 code": {
|
||||
"estimatedMillionTokens": 94.16,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"hy4 preview": {
|
||||
"estimatedMillionTokens": 71.07,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"gpt-5.6 luna": {
|
||||
"estimatedMillionTokens": 109.57,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"minimax m3": {
|
||||
"estimatedMillionTokens": 302.7,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"mimo-v2.5-pro": {
|
||||
"estimatedMillionTokens": 80.46,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"qwen 3.6 plus": {
|
||||
"estimatedMillionTokens": 175.3,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"minimax m2.7": {
|
||||
"estimatedMillionTokens": 302.7,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"minimax m2.5": {
|
||||
"estimatedMillionTokens": 302.7,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"qwen 3.7 plus": {
|
||||
"estimatedMillionTokens": 254.55,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"hy3": {
|
||||
"estimatedMillionTokens": 609.43,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"qwen 3.8 flash": {
|
||||
"estimatedMillionTokens": 357.95,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"glm-5.3 flash": {
|
||||
"estimatedMillionTokens": 635.96,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"deepseek v4 flash vision exp": {
|
||||
"estimatedMillionTokens": 187.92,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"longcat-2.0": {
|
||||
"estimatedMillionTokens": 375.84,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"deepseek v4 flash": {
|
||||
"estimatedMillionTokens": 375.84,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"deepseek v4.1 flash": {
|
||||
"estimatedMillionTokens": 751.68,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"mimo-v2.5": {
|
||||
"estimatedMillionTokens": 980.93,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"muse spark 1.3 contributor": {
|
||||
"estimatedMillionTokens": 1373.3,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
},
|
||||
"muse spark 1.2 contributor": {
|
||||
"estimatedMillionTokens": 1373.3,
|
||||
"basis": "list-price-credit",
|
||||
"confidence": "high"
|
||||
}
|
||||
},
|
||||
"notes": "New Sept 2026 low-cost subscription tier"
|
||||
"estimatedTokenBudget": {
|
||||
"description": "$10/mo open-model subscription ($15–$60 per-model allowances; ~30M–752M tokens)",
|
||||
"estimatedMillionTokens": 30,
|
||||
"midpointEstimate": 95,
|
||||
"optimisticEstimate": 752,
|
||||
"assumptions": "$10/mo plan fee provides independent monthly usage allowances per model ($15, $30, or $60) with 20% 5h and 50% weekly caps; tokens range from 7.85M (Kimi K3 at $15 allowance) to 94M (Kimi K2.7 Code at $60) up to 752M (DeepSeek V4.1 Flash at $60 promo) and 981M (MiMo-V2.5)",
|
||||
"estimateMeta": {
|
||||
"sourceUrl": "https://opencode.ai/docs/go/#usage-limits",
|
||||
"sourceQuote": "Usage limits are defined as monthly dollar amounts... Each model has the following usage limits: 5-hour — 20% of the monthly limit; weekly — 50%; and monthly — 100%. For example, if a model has a $60 monthly limit, you can spend up to: 5-hour limit — $12, weekly limit — $30, monthly limit — $60.",
|
||||
"sourceType": "official",
|
||||
"confidence": "high",
|
||||
"verifiedAt": "2026-09-19",
|
||||
"basisModel": "open-model-allowances"
|
||||
}
|
||||
},
|
||||
"notes": "New low-cost open-model subscription with independent allowances per model"
|
||||
}
|
||||
],
|
||||
"gotchas": [
|
||||
"The CLI itself is free/open-source; opencode.ai/pricing is retired (404) - offerings are Zen (PAYG) and Go ($10/mo)",
|
||||
"Zen charges $1.23 card processing fee on top-ups; auto top-up at $20 below $5 balance",
|
||||
"GPT-5.6 Sol carries 50% discount through Sept 18, 2026",
|
||||
"Zen models hosted in US with zero-retention provider policy"
|
||||
"OpenCode Go ($10/mo) provides $15, $30, or $60 monthly usage allowance per model with 20% 5h and 50% weekly caps",
|
||||
"Optional 'Use balance' falls back to Zen PAYG balance when Go model limits are reached",
|
||||
"DeepSeek V4.1 Flash has 4× promotional limits ($60/mo limit) through Sep 20, 2026 (normally $15)",
|
||||
"Muse Spark Contributor models train on prompt/output data and are not zero data retention (ZDR); all other Go models provide zero training (0 days retention, except Grok 4.6 and GPT-5.6 Luna with 30 days retention)",
|
||||
"Zen PAYG charges $1.23 card processing fee on top-ups; auto top-up at $20 below $5 balance"
|
||||
],
|
||||
"tosHighlights": [],
|
||||
"dataTraining": "Zero-retention; no training on Zen",
|
||||
"tosHighlights": [
|
||||
"Zero data retention on most Go models (0 days); 30 days on Grok 4.6 and GPT-5.6 Luna; Muse Spark Contributor trains on code",
|
||||
"Zen models US-hosted with zero retention"
|
||||
],
|
||||
"dataTraining": "Zero data retention (0 days) on most models; 30 days on Grok 4.6 & GPT-5.6 Luna; Muse Spark Contributor trains on code",
|
||||
"ipIndemnity": false,
|
||||
"stackingPolicy": "prohibited",
|
||||
"stackingPolicyNote": "Terms of Service: prohibits users who \"create, maintain, or use multiple accounts to circumvent usage limits, access restrictions, billing obligations... or any other restriction or policy.\""
|
||||
|
||||
+1323
-166
File diff suppressed because it is too large
Load Diff
Reference in New Issue
Block a user